AMD CPU speculation... and expert conjecture

[bobbybamf12]

The growth in speed increases are outpacing Intels.
Its been shown here that AMDs "IPC" if you will, isnt as bad as some thought, and compared to BD, PD showed huge gains.
SR hints ad another set of nice gains, again out distancing Intels growth, and lets just look at the first APU iterations, where no clock speeds, no tweaks, using the old VLIW approach, it was miles ahead, and being held back by BW, where the competition wasnt even fast enough to take advantage of the BW it had.
Upping it to GDDR5, with its BW, using GCN 1.1 if you will, improving clocks and power, then improving power management where AMD has been further behind Intel, we again will start seeing faster growth here as well, allowing for a more sustainable turbo, which Intel already enjoys, and is why unless these chips are locked in at a given frequency, so far, this advantage has been huge for Intel, where the growth is less from them, as we see HSW adding a better gfx solution, and in house VRMs, but the TDP is up in high usage, and the "IPC", if you will, is very flat.
So, AMDs gfx solutions, huge upside.
CPU, nice upside.
Power management, huge upside there as well, and until we see the full growth of the BD design, which we wont until EX, where perf is supposed to skyrocket, I cant see why people wont be picking up these chips, and future ones as well, which is alot more than can be said before RR, where they let BD out before it was truly ready, and why we see the current downtrend, but they arent PDs, or SRs, and these are right around the corner, in varying iterations

It's pretty obvious that you're an AMD fanboy, but no worries, as I am actually a bit biased towards AMD myself. But to blindly close your eyes and not see the truth is wrong.

Your 15-20% faster is more like the lower limit, not the upper one!! As i've just shown with 2 calculations, in which Intel is 37% and 50% faster. Its a reproducible calculation with most single threaded tests, and you can calculate them yourself when you have time.

There's nothing wrong with having subpar single core performance, especially when you can more than make up for it in multi threaded workloads like AMD does.
For example, encoding a song is only a matter of seconds, and is irrelevant, but when re-encoding videos with x264 for example, AMD usually provides much more performance per $, and that I like. The i3 vs FX 6300 is an outshining example of this

The issues lie in your "IPC" calculations...you cannot deduce IPC from anything anyone has posted...it's not feasibly possible. The limits for IPC are hard coded into the architecture because they are hardware limited. None of the numbers are anywhere near what some of you are talking about either. IPC over 40 is a bit crazy...yes, it is dictated by program's coding, but they never exceed the engineered maximum...that's why overclocking has become so popular...you can't increase IPC, but you can increase the clock cycles per second.

IPC has little relevance from software...the term is thrown about far too often. IPC is an architecture term from engineering as the maximum theoretical capabilities of the hardware itself. Somehow or another, it got attached to software, and it's actually nothing to do with software.

Coding efficiency and optimization of architecture are the advantages Intel has over AMD...those are the only advantages...

If you look at them on paper...the AMD is easily a LOT more CPU...

The issue is the hardware is not being advancely optimized by the coders...once that occurs...look out.

Intel also has better power consumption and those two advantages are HUGE advantages.

 

[8350rocks]

The growth in speed increases are outpacing Intels.
Its been shown here that AMDs "IPC" if you will, isnt as bad as some thought, and compared to BD, PD showed huge gains.
SR hints ad another set of nice gains, again out distancing Intels growth, and lets just look at the first APU iterations, where no clock speeds, no tweaks, using the old VLIW approach, it was miles ahead, and being held back by BW, where the competition wasnt even fast enough to take advantage of the BW it had.
Upping it to GDDR5, with its BW, using GCN 1.1 if you will, improving clocks and power, then improving power management where AMD has been further behind Intel, we again will start seeing faster growth here as well, allowing for a more sustainable turbo, which Intel already enjoys, and is why unless these chips are locked in at a given frequency, so far, this advantage has been huge for Intel, where the growth is less from them, as we see HSW adding a better gfx solution, and in house VRMs, but the TDP is up in high usage, and the "IPC", if you will, is very flat.
So, AMDs gfx solutions, huge upside.
CPU, nice upside.
Power management, huge upside there as well, and until we see the full growth of the BD design, which we wont until EX, where perf is supposed to skyrocket, I cant see why people wont be picking up these chips, and future ones as well, which is alot more than can be said before RR, where they let BD out before it was truly ready, and why we see the current downtrend, but they arent PDs, or SRs, and these are right around the corner, in varying iterations

It's pretty obvious that you're an AMD fanboy, but no worries, as I am actually a bit biased towards AMD myself. But to blindly close your eyes and not see the truth is wrong.

Your 15-20% faster is more like the lower limit, not the upper one!! As i've just shown with 2 calculations, in which Intel is 37% and 50% faster. Its a reproducible calculation with most single threaded tests, and you can calculate them yourself when you have time.

There's nothing wrong with having subpar single core performance, especially when you can more than make up for it in multi threaded workloads like AMD does.
For example, encoding a song is only a matter of seconds, and is irrelevant, but when re-encoding videos with x264 for example, AMD usually provides much more performance per $, and that I like. The i3 vs FX 6300 is an outshining example of this

The issues lie in your "IPC" calculations...you cannot deduce IPC from anything anyone has posted...it's not feasibly possible. The limits for IPC are hard coded into the architecture because they are hardware limited. None of the numbers are anywhere near what some of you are talking about either. IPC over 40 is a bit crazy...yes, it is dictated by program's coding, but they never exceed the engineered maximum...that's why overclocking has become so popular...you can't increase IPC, but you can increase the clock cycles per second.

IPC has little relevance from software...the term is thrown about far too often. IPC is an architecture term from engineering as the maximum theoretical capabilities of the hardware itself. Somehow or another, it got attached to software, and it's actually nothing to do with software.

Coding efficiency and optimization of architecture are the advantages Intel has over AMD...those are the only advantages...

If you look at them on paper...the AMD is easily a LOT more CPU...

The issue is the hardware is not being advancely optimized by the coders...once that occurs...look out.

Intel also has better power consumption and those two advantages are HUGE advantages.

$3-5 cheaper to run per year is not a HUGE advantage...that's just a myth. You would literally have to run the i7-3770k for 20+ years to make up the initial cost difference in electric bill savings.

 

[griptwister]

The only way how power consumption would be a total advantage is in mobile computing. People are total idiots when they say the FX 8350 has bad power consumption. Keep in mind people, It's an 8 core CPU. Intel hexicore CPUs have 130w requirement for boards correct? Imo, I think the FX 8350 is pretty decent due to how many cores it has.

And there is the thought that having less power draw means you can buy a cheaper PSU... but most people that buy a "High end" cpu can afford a big powersupply.

@Cazalan, So, is it a possibility that AMD has too much L2?

 

[8350rocks]

From what I've seen, a BD/PD MODULE is about 75-80% the performance of a Intel core, at the same clock. AMD clocks more agressively though...

Considering the average of a number of workloads, I'd almost agree with that figure, but I think 65-75% is a more accurate range

Higher clock speeds work fine on the desktop. Laptops are where AMD needs to do something. Was looking at a few laptops the other day, and man the A8 and A10 really do need to improve their clocks. 1.9Ghz is really a joke. Plus, I've read somewhere that these APU's can't sustain their max turbo state for more than a few seconds. They really look weak against even a non turboing mobile i3, let alone the i5's.

I really hate to burst your bubble; however, in your earlier post...you equated AMD to 66% performance...

However, your fractional math to get to 1.51% was off...

80 is 1.33 times 60...

If you take 80/60=1.33 this makes intel's PPS performance only 133% of AMD's and in a skewed comparison that shows an artificially higher number to begin with...I think it is very fair to say reality falls somewhere around 15-25% as was stated before...

That was a great job selling the sensationalism of your flawed mathematics though, you had the intel crowd roaring about supremacy by 50% for nearly 3 pages.

As I said earlier, the difference is not that far off, and if you take real world numbers and exclude theoretical engineered maximums 1 AMD module is equal to approximately 1.6 intel cores

@griptwister:

The AMD mobile solutions do not exceed 35W TDP and draw low voltage.

 

[cowboy44mag]

I've really been enjoying this thread, and think I have actually been learning quite a bit here and there. I'm not going to come on here and talk about a bunch of theoretical calculations, benchmarks and such, I'm going to talk about the real world. Every Intel fan boy going will tell you my PC cant possibly be a "gaming computer" because I "only" run an AMD Phenom II 965 BE processor. I have had a bunch of "expert" Intel fan boys tell me that an i3 is a super computer compared to my computer and have heard a few of them even say that for gaming an i3 can beat anything AMD has.

To these people I only have to ask one thing: Are you really that stupid? I hate to tell you but my computer can out game an i3, and totally dominate it in multitasking. Now I'm not talking about old games made expressly to run on 2 cores, but the new games that are made to run on 3+ cores, and future games that will be made to run on more than 4 cores. Intel fan boys, the i3 (for gaming) is DEAD technology because dual core gaming is a thing of the past. I just bought Bioshock Infinite a week ago and it recommends a quad core processor, soon games will require a quad core processor, the writing is on the wall. My lowly 965 BE (clocked at 4.1Ghz for everyday use, but I have hit 4.5 stable) plays Skyrim, Crysis 3, Far Cry 3, Bioshock Infinite and a host of other games at ultra settings. That is real world use. I know that Piledriver must be amazing to game with as it is a big improvement over the Phenom, and my Sabertooth 990FX is hungry for Steamroller as it is going to be amazing.

Plain and simple if you have more money than you know what to do with then by all means buy a high end i7, be an Intel fan boy and keep printouts of your benchmark scores on your nightstand next to your lube and Kleenex. If you want a computer that can do it all except rule synthetic benchmark test than there is nothing wrong with a good quality AMD system.

 

[anxiousinfusion]

Are you really that stupid?

I just see a bunch of i3 gamers trying to justify their purchase. It must be difficult to have spent your only ~400 dollars on a gaming build only to constantly hear online how much better everyone else's i5+ builds are at gaming. "Quick! I better go bash other people's hardware to make myself feel better about my mediocre computer!"

They're not stupid... just human.

 

[8350rocks]

I've really been enjoying this thread, and think I have actually been learning quite a bit here and there. I'm not going to come on here and talk about a bunch of theoretical calculations, benchmarks and such, I'm going to talk about the real world. Every Intel fan boy going will tell you my PC cant possibly be a "gaming computer" because I "only" run an AMD Phenom II 965 BE processor. I have had a bunch of "expert" Intel fan boys tell me that an i3 is a super computer compared to my computer and have heard a few of them even say that for gaming an i3 can beat anything AMD has.

To these people I only have to ask one thing: Are you really that stupid? I hate to tell you but my computer can out game an i3, and totally dominate it in multitasking. Now I'm not talking about old games made expressly to run on 2 cores, but the new games that are made to run on 3+ cores, and future games that will be made to run on more than 4 cores. Intel fan boys, the i3 (for gaming) is DEAD technology because dual core gaming is a thing of the past. I just bought Bioshock Infinite a week ago and it recommends a quad core processor, soon games will require a quad core processor, the writing is on the wall. My lowly 965 BE (clocked at 4.1Ghz for everyday use, but I have hit 4.5 stable) plays Skyrim, Crysis 3, Far Cry 3, Bioshock Infinite and a host of other games at ultra settings. That is real world use. I know that Piledriver must be amazing to game with as it is a big improvement over the Phenom, and my Sabertooth 990FX is hungry for Steamroller as it is going to be amazing.

Plain and simple if you have more money than you know what to do with then by all means buy a high end i7, be an Intel fan boy and keep printouts of your benchmark scores on your nightstand next to your lube and Kleenex. If you want a computer that can do it all except rule synthetic benchmark test than there is nothing wrong with a good quality AMD system.

1.) Welcome to the thread!

2.) Nice name...that is an awesome handgun/leveraction cartridge...

3.) I think the i3 owners feel like their "e-peen" is too small to measure up against relevant competition, so they try to measure against last generation technology, that is honestly still superior; but it makes them feel "worthy" because their "e-peen" almost measures up to something that was quite good a few years ago.

 

[palladin9479]

I feel, If AMD would give the APUs some gosh darned L3 Cache, they'd be up there with intel. I'd like to buy a new CPU sometime after july, probably Kaveri, and I'd like to upgrade to a CPU with L3 cache. Lol, I'm not sure why they can't/don't do this, would it have a slightly higher power footprint? I really want to upgrade this Phenom II x4 840, but I don't want to waste money. (Cpu has no L3 ):

Then it would no longer be an APU.

The GPU component of an APU is something in the neighborhood of ~50% of the die real estate. To make a APU with decent CPU power while also being low cost means that ~something~ had to be cut out. Of all the components on a CPU the one that has the least performance to space yield is the L3 cache. Memory arrays take up large amounts of space, no way around it. So the logical component to cut out of the die and replace with a GPU is the L3, and that's exactly what AMD did.

L3 in AMD systems really doesn't help much, it's more or less just a slow buffer between the CPU and main memory. L1 & L2 are by far the most important components, which is why you see erratic performance with PD/BD due to their wonky caching access mechanism and shared backend L2 cache. It's been demonstrated that if you disable one core per module the CPU's per core performance jumps up significantly, obviously not SB levels due to 2 ALU vs 3 ALUs and Intel's superior branch prediction / caching system but much higher then what BD/PD gets natively. L3 in Intel's is a bit different as Intel's caching is inclusive and the shared L3 needs to hold the contents of all the L2 plus extra. IB is only 256KB per core or 1024KB total per CPU for L2. Llano and beyond is 1MB per core with BD/PD having 8MB per CPU. So you can see that BD/PD has 4x the amount of L2 cache per core but that cache is slower and has a less efficient access mechanism / buffer then SB.

 

[cowboy44mag]

I've really been enjoying this thread, and think I have actually been learning quite a bit here and there. I'm not going to come on here and talk about a bunch of theoretical calculations, benchmarks and such, I'm going to talk about the real world. Every Intel fan boy going will tell you my PC cant possibly be a "gaming computer" because I "only" run an AMD Phenom II 965 BE processor. I have had a bunch of "expert" Intel fan boys tell me that an i3 is a super computer compared to my computer and have heard a few of them even say that for gaming an i3 can beat anything AMD has.

To these people I only have to ask one thing: Are you really that stupid? I hate to tell you but my computer can out game an i3, and totally dominate it in multitasking. Now I'm not talking about old games made expressly to run on 2 cores, but the new games that are made to run on 3+ cores, and future games that will be made to run on more than 4 cores. Intel fan boys, the i3 (for gaming) is DEAD technology because dual core gaming is a thing of the past. I just bought Bioshock Infinite a week ago and it recommends a quad core processor, soon games will require a quad core processor, the writing is on the wall. My lowly 965 BE (clocked at 4.1Ghz for everyday use, but I have hit 4.5 stable) plays Skyrim, Crysis 3, Far Cry 3, Bioshock Infinite and a host of other games at ultra settings. That is real world use. I know that Piledriver must be amazing to game with as it is a big improvement over the Phenom, and my Sabertooth 990FX is hungry for Steamroller as it is going to be amazing.

Plain and simple if you have more money than you know what to do with then by all means buy a high end i7, be an Intel fan boy and keep printouts of your benchmark scores on your nightstand next to your lube and Kleenex. If you want a computer that can do it all except rule synthetic benchmark test than there is nothing wrong with a good quality AMD system.

1.) Welcome to the thread!

2.) Nice name...that is an awesome handgun/leveraction cartridge...

3.) I think the i3 owners feel like their "e-peen" is too small to measure up against relevant competition, so they try to measure against last generation technology, that is honestly still superior; but it makes them feel "worthy" because their "e-peen" almost measures up to something that was quite good a few years ago.

Thanks for the welcome. I'm not an expert on processors, my area of expertise is horses actually. I own a horse boarding business and give cowboy action shooting lessons. I got to admit that I love the .44, be it cowboy load or full boar magnum. I actually run my shooting course using a Ruger Super Blackhawk, my personal favorite revolver. But back to topic....

Too many people (Intel fan boys) dismiss AMD as "garbage" and "non competition" when AMD is right in the game. People don't realize that even "older" last generation processors like my Phenom II 965 BE can overclock well and are still playing the newest games out at the highest settings possible.

I think when it comes down to it most people build their system for a practical reason, I built mine primarily for advertising my business and keeping track of all the books, then we all come back to gaming. Gaming is where the "great divide" comes from with the whole myth of Intel superiority. I call it a myth because the "superiority" is usually measured at 5-10 fps (max difference I've seen in real world gaming is 20 fps) which most of the time the human eye isn't even going to notice. Its a superiority based on synthetic benchmarks a few fps and mere seconds. But that superiority is being cut by less and less every time AMD releases a new processor.

I wonder with all the speculation on single core performance how important single core performance is going to be in the new AMD dominated world of video games. Lets face it Intel dropped the ball and AMD got the contracts from Sony and Microsoft for the next gen high end console systems. What that means is that studios will be producing games optimized to run on AMD architecture, and that means heavily multi-threaded video games are going to be what the future holds. If games go from needing less single core performance to more multi-core performance where would that leave Intel in the gaming world? All Intel fan boys tout you don't need 8 cores you need 4 powerful cores, what if that were no longer true? Right now yea you still rule the all-important benchmarks, you still play games with 10 more fps than AMD players get, I would recommend enjoying it while you can.

What would happen right now if a Piledriver and an i7 both tried to play a heavily multi-threaded game made to run on 8 cores? Where would Intel's "vast superiority" be? Now while still thinking about this scenario.... the i7 is having trouble outperforming the Piledriver, Haswell isn't a major upgrade over Ivy Bridge, and here comes Steamroller!! I'm just hoping the first gen Steamroller processors are still going to AM3+!! I would love to see just what this Sabertooth 990FX could do with a Steamroller cpu!!

 

[-Fran-]

I've really been enjoying this thread, and think I have actually been learning quite a bit here and there. I'm not going to come on here and talk about a bunch of theoretical calculations, benchmarks and such, I'm going to talk about the real world. Every Intel fan boy going will tell you my PC cant possibly be a "gaming computer" because I "only" run an AMD Phenom II 965 BE processor. I have had a bunch of "expert" Intel fan boys tell me that an i3 is a super computer compared to my computer and have heard a few of them even say that for gaming an i3 can beat anything AMD has.

To these people I only have to ask one thing: Are you really that stupid? I hate to tell you but my computer can out game an i3, and totally dominate it in multitasking. Now I'm not talking about old games made expressly to run on 2 cores, but the new games that are made to run on 3+ cores, and future games that will be made to run on more than 4 cores. Intel fan boys, the i3 (for gaming) is DEAD technology because dual core gaming is a thing of the past. I just bought Bioshock Infinite a week ago and it recommends a quad core processor, soon games will require a quad core processor, the writing is on the wall. My lowly 965 BE (clocked at 4.1Ghz for everyday use, but I have hit 4.5 stable) plays Skyrim, Crysis 3, Far Cry 3, Bioshock Infinite and a host of other games at ultra settings. That is real world use. I know that Piledriver must be amazing to game with as it is a big improvement over the Phenom, and my Sabertooth 990FX is hungry for Steamroller as it is going to be amazing.

Plain and simple if you have more money than you know what to do with then by all means buy a high end i7, be an Intel fan boy and keep printouts of your benchmark scores on your nightstand next to your lube and Kleenex. If you want a computer that can do it all except rule synthetic benchmark test than there is nothing wrong with a good quality AMD system.

1.) Welcome to the thread!

2.) Nice name...that is an awesome handgun/leveraction cartridge...

3.) I think the i3 owners feel like their "e-peen" is too small to measure up against relevant competition, so they try to measure against last generation technology, that is honestly still superior; but it makes them feel "worthy" because their "e-peen" almost measures up to something that was quite good a few years ago.

Thanks for the welcome. I'm not an expert on processors, my area of expertise is horses actually. I own a horse boarding business and give cowboy action shooting lessons. I got to admit that I love the .44, be it cowboy load or full boar magnum. I actually run my shooting course using a Ruger Super Blackhawk, my personal favorite revolver. But back to topic....

Too many people (Intel fan boys) dismiss AMD as "garbage" and "non competition" when AMD is right in the game. People don't realize that even "older" last generation processors like my Phenom II 965 BE can overclock well and are still playing the newest games out at the highest settings possible.

I think when it comes down to it most people build their system for a practical reason, I built mine primarily for advertising my business and keeping track of all the books, then we all come back to gaming. Gaming is where the "great divide" comes from with the whole myth of Intel superiority. I call it a myth because the "superiority" is usually measured at 5-10 fps (max difference I've seen in real world gaming is 20 fps) which most of the time the human eye isn't even going to notice. Its a superiority based on synthetic benchmarks a few fps and mere seconds. But that superiority is being cut by less and less every time AMD releases a new processor.

I wonder with all the speculation on single core performance how important single core performance is going to be in the new AMD dominated world of video games. Lets face it Intel dropped the ball and AMD got the contracts from Sony and Microsoft for the next gen high end console systems. What that means is that studios will be producing games optimized to run on AMD architecture, and that means heavily multi-threaded video games are going to be what the future holds. If games go from needing less single core performance to more multi-core performance where would that leave Intel in the gaming world? All Intel fan boys tout you don't need 8 cores you need 4 powerful cores, what if that were no longer true? Right now yea you still rule the all-important benchmarks, you still play games with 10 more fps than AMD players get, I would recommend enjoying it while you can.

What would happen right now if a Piledriver and an i7 both tried to play a heavily multi-threaded game made to run on 8 cores? Where would Intel's "vast superiority" be? Now while still thinking about this scenario.... the i7 is having trouble outperforming the Piledriver, Haswell isn't a major upgrade over Ivy Bridge, and here comes Steamroller!! I'm just hoping the first gen Steamroller processors are still going to AM3+!! I would love to see just what this Sabertooth 990FX could do with a Steamroller cpu!!

I come from a 965BE at 3.9Ghz and jumped from AM3+ after having microcenter offer me a 2700k for 180USD and an Asus Z77 deluxe board for 100USD, haha. I was waiting for the 8350, but they didn't arrive before me leaving town.

Since I did the switch, the games almost doubled in FPS'es (mostly UE3 engine games and Valve games) using the same GTX670, the TT FRIO, same RAM (4x4GB DDR3 1600 C8) and the i7 at 4.5Ghz.

It's all a matter of price. Always. A FX4300 priced in the same slot as any locked i3 and i5, the FX is the better buy; even more, a few more bucks get you a FX6300, but IMO better get a decent cooler (Hyper 212 + FX4300) than just the FX6300 + stock HSF (sucks so bad...) to squeeze a little more life of the FX4300. That's in the "low/mid end" bracket. MoBo wise, it's a more mixed bag. The AMD970 (non-G) chipset MoBos are not that expensive and are way better than H6x MoBos from Intel (albeit a little more pricey). I just found 1 Z77 from MSI that was in the same bracket price as a 970, but no 95W+ CPU support; most 970s support 125W CPUs. Weird thing is in APU territory; most MoBos are a bit pricey for my taste, but they do offer a lot of goodies.

Uhm... I think I went a little sideways to the discussion, but my bottom line is that in the mid and low, things are moved towards AMD. The Pentium G8x0 is no longer a viable gaming CPU, so the APUs (not Llano though; Trinity+) shine in the low, plus the mid being dominated by the FX4300 (IMO) and in the upper by the i5-K or FX83x0, where the FX has more multithreading for it's price, but the i5 has more muscle per clock (in my experience) which favor games. If you have the extra cash for an i7, then get a FX8350 + a good HSF.

I think the simple impressions in life are the best. AMD is not far behind Intel; that's a good thing, but let's not obscure the fact that for a given price, AMD can be in fact a better buy. Those are my impressions after playing with a FX4300 and an i3 side by side. And as a side note; minecraft is a really good benchmark. make a circuit to start spawning mobs and watch the frames drop

Cheers!

 

[jdwii]

If Intel was 51% better than AMD...then how on earth does the FX8350 beat Intel's i5-3570k and i7-3770k at so many things? Is it just 175% better at certain things? How do you close the loop for your logic?

Your math is flawed...and so is your logic.

The FX-8350 has a bottleneck in the instruction decoding, so it can't keep the pipes full. The backend actually has more execution resources which is why fully parallelized tasks can beat the i5/i7 quads in many things.

This is why SR is going to dual instruction decoders, which will clear up that bottleneck (hopefully) and provide more consistent performance.

I have seen the logic flowcharts for steamroller, and it looks incredibly promising...they predict 30% greater efficiency in mispredicted branches, and combine that with shorter pipelines, decreased memory latency, etc. It will be a giant leap forward...I equate it to the leap that occured at K7/K8 in terms of hardware advancement percentage gains.

I know this will sound bad but why didn't Amd do this with Bulldozer why wait and get all the bad rep, was Amd's engineers to blame for the mistake? Even us tomshardware readers could easily see the bad decisions in the design.

 

[$hawn]

Based on GFLOPS, 1 AMD MODULE is worth 2.2 intel cores....or an intel core is worth about 45% of 1 AMD MODULE (64/28 = 2.2xxx)

Please refer to the manufacturer's maximum GFLOPS for single and double precision:

i7-3770k SP GFLOPS: 112
i7-3770k DP GFLOPS: 28
FX8350 SP GFLOPS: 256
FX8350 DP GFLOPS: 64

“1 AMD MODULE is worth 2.2 intel cores” --> Very interesting that. I’d like to see atleast 3 to 4 real world benchmarks that prove the same. Say a 4C/4T core i5 (no HT) verses an FX 4300 (4C/2M), with both at either the same clock speed, or the same power consumption( your choice).

I’d like to know where you got those numbers from. Are these practically achieved values, or theoretical ones?
You seem to be basing your 2.2x claim based on Floating point calculation ability, not on much more frequently used integer crunching power, or even on a more realistic mix of integer and floating point calculation ability. Can you understand why AMD decided to share the FPU between 2 cores in a module?

I really hate to burst your bubble; however, in your earlier post...you equated AMD to 66% performance...

Its really annoying when people like you and noob, read a person’s statement in half, and then start jumping around without even understanding what it means!!

The amd being only 66% of an intel core value (ie, intel being 50% faster) was calculated for AAC encoding. To the best of my knowledge, for AAC encoding, this is a true value, and I STILL stand by it.
I challenge you to prove to me that the IPG ratio of Intel SB(or IB) , verses AMD PD, for AAC encoding, is less than 1.50 by a significant margin, atleast around 1.33, as per your claims. Can you??

Similarly, I calculated the Intel vs AMD IPG ratio for Cinebench (singlethreaded) to be 1.37. Again,prove to me that it’s less than that by a significant margin; prove that it’s around 1.20 and I will gladly accept that my calculations were flawed.

Please do patiently read my previous long reply to noob a few posts above. I think it will help greatly in clearing your confusion.

The 65-75% value I’ve stated to gamerk, was my guess on the average IPG ratios of AMD vs Intel, over a number of different workloads.

That was a great job selling the sensationalism of your flawed mathematics though, you had the intel crowd roaring about supremacy by 50% for nearly 3 pages.

It’s funny how I get alternatively accused of being an AMD fanboy, and then an Intel on now. For the past 3 months, I’ve been called an AMD fanboy for aggressively supporting an FX 6300 over an i3, and an FX 8350 over an i5-3470.

In the end, this ratio debate means nothing in the real world, where things like higher clock speeds, more cores and relatively powerful iGPUs help AMD make up for it. Nobody sits and encodes AAC on his/her system for the whole day AFAIK!! But in the end, fact remains that AMD is still behind in IPG, by a more than significant margin.

 

[jdwii]

cowboy44mag

Ask them how they feel about their I3's lacking the ability of overclocking which is something a gaming computer should be able to do IMHO. Also ask them how they feel about Intel making people charge extra money for performance they already have on their CPU http://www.engadget.com/2011/08/15/the-intel-upgrade-service-once-again-charging-you-50-to-do-stu/ Unlike Amd where its a simple option in the Bios i even think Intel threten Board makers not to do it.

Don't forget about how much better ivy was over sandy in overclocking or how haswell will be a tock instead of a tick based on Intel's Tick tock strategy even though tomshardware benchmarks show some disappointment. Sandy(Tock)Ivy (tick) Haswell (Tock).

And if all else fails talk about their so called 650 GTX video card performance from haswell which somehow fails to beat Trinity's IGPU.

Not to mention they make their I7's Cost more when they just have HT and i don't even think it cost more to make i think its the same size die.

Although i really question Amd and their decisions Intel still makes me a little more disappointed. I mean we heard so much from people expecting a lot from haswell IGPU and really its not that amazing.

 

[$hawn]

Higher clock speeds work fine on the desktop. Laptops are where AMD needs to do something. Was looking at a few laptops the other day, and man the A8 and A10 really do need to improve their clocks. 1.9Ghz is really a joke. Plus, I've read somewhere that these APU's can't sustain their max turbo state for more than a few seconds. They really look weak against even a non turboing mobile i3, let alone the i5's.

Were they the old Llano models? The newer Trinity models run faster.

http://www.staples.com/HP-Pavilion-G6-2235us-156-Laptop/product_985050

A6-4400M is 2.7/3.2GHz - not bad for $399

No, they were all trinity A8's and A10's. I stay in Bangalore, and in India stuff costs around 10-20% more than in the US. You can look up the models I was interested in at flipkart.com.
HP G6 2314AX and HP 2313AX. The i3 one was HP G6-2231TX

 

[jdwii]

$hawn just don't listen to them i got called the same when really i dislike both even though i dislike Intel a bit more.

I'd even say a module is bigger than 2.2 times a single core from Intel before i say its 2.2 times as good as 2 Intel cores that's just amazing maybe from some synthetic benchmark. I think its safe to say Piledrivers design is around 33% slower than Ivy's design per core per clock.

Amd's design uses more die size for less performance even when Intel was at 32nm. There is no other way to look at that but the design being inefficient compared to their competitor not only does this raise heat but it also lowers profits for Amd and makes their future development or research suffer a bit.

 

[aggroboy]

Is the 28nm process a big handicap versus Intel's 22nm?

 

[palladin9479]

If Intel was 51% better than AMD...then how on earth does the FX8350 beat Intel's i5-3570k and i7-3770k at so many things? Is it just 175% better at certain things? How do you close the loop for your logic?

Your math is flawed...and so is your logic.

The FX-8350 has a bottleneck in the instruction decoding, so it can't keep the pipes full. The backend actually has more execution resources which is why fully parallelized tasks can beat the i5/i7 quads in many things.

This is why SR is going to dual instruction decoders, which will clear up that bottleneck (hopefully) and provide more consistent performance.

I have seen the logic flowcharts for steamroller, and it looks incredibly promising...they predict 30% greater efficiency in mispredicted branches, and combine that with shorter pipelines, decreased memory latency, etc. It will be a giant leap forward...I equate it to the leap that occured at K7/K8 in terms of hardware advancement percentage gains.

I know this will sound bad but why didn't Amd do this with Bulldozer why wait and get all the bad rep, was Amd's engineers to blame for the mistake? Even us tomshardware readers could easily see the bad decisions in the design.

Die space most likely. Those probably though that it would work out and didn't predict software being as diverse and generic as it is. Basically law of unintended consequences being what it is. If their die shrinking then that frees up some space and they'll be using it to create another scheduler and a larger cache buffer / ect.

Direct comparison between PD and IB is kinda hard to do. IB has 3 ALU's per "core" while the PD only has 2 ALU's per "core" so immediately we're talking 33% less theoretical maximum integer performance per core. The FPU's capacity are comparable per core though the Intel's units are far more efficient due to their multiple pipeline design. When we're talking per chip your looking at 16 total ALU's for the eight "core" AMD PD chip and 12 total ALU's for the four "core" IB chip. So 33% more total theoretical integer performance for the AMD chip and about double the theoretical FPU performance though you won't ever see it hit maximum.

So for raw processing capacity the 8350 beats out any four core intel can produce while the Intel six core beats out the AMD part. Now actual realized capacity is a different matter entirely, BD/PD's relative newness combined with AMD's significantly lower R&D budget creates a handicap for the design that must be overcome. Intel's design is far more mature and developed, and we can see that in many benchmarks.

 

[8350rocks]

“1 AMD MODULE is worth 2.2 intel cores” --> Very interesting that. I’d like to see atleast 3 to 4 real world benchmarks that prove the same. Say a 4C/4T core i5 (no HT) verses an FX 4300 (4C/2M), with both at either the same clock speed, or the same power consumption( your choice).

I’d like to know where you got those numbers from. Are these practically achieved values, or theoretical ones?
You seem to be basing your 2.2x claim based on Floating point calculation ability, not on much more frequently used integer crunching power, or even on a more realistic mix of integer and floating point calculation ability. Can you understand why AMD decided to share the FPU between 2 cores in a module?

Those are manufacturer stated theoretical maximums of the hardware taken from AMD and Intel hardware specs...(respectively). I chose the comparison because intel used to quote GLFOPS for their CPUs when they had the distinct advantage...(note they no longer do this for a reason)

Its really annoying when people like you and noob, read a person’s statement in half, and then start jumping around without even understanding what it means!!

The amd being only 66% of an intel core value (ie, intel being 50% faster) was calculated for AAC encoding. To the best of my knowledge, for AAC encoding, this is a true value, and I STILL stand by it.
I challenge you to prove to me that the IPG ratio of Intel SB(or IB) , verses AMD PD, for AAC encoding, is less than 1.50 by a significant margin, atleast around 1.33, as per your claims. Can you??

Similarly, I calculated the Intel vs AMD IPG ratio for Cinebench (singlethreaded) to be 1.37. Again,prove to me that it’s less than that by a significant margin; prove that it’s around 1.20 and I will gladly accept that my calculations were flawed.

Please do patiently read my previous long reply to noob a few posts above. I think it will help greatly in clearing your confusion.

The 65-75% value I’ve stated to gamerk, was my guess on the average IPG ratios of AMD vs Intel, over a number of different workloads.

Ok, let's compare a wide range of benchmarks, shall we:

http://www.xbitlabs.com/articles/cpu/display/fx-8350-8320-6300-4300_7.html

WinRAR 4.2:
FX8350: 68 seconds
i7-3770k: 57 seconds (+19%)
i5-3570k: 83 seconds

TrueCrypt 7.1a
FX 8350: 259 MB/s (+13%)
i7-3770k: 229 MB/s
i5-3570k: 171 MB/s

iTunes:
FX8350: 185.1 seconds
i7-3770k: 126.8 sec (+49%)
i5-3570k: 130.2 sec

Adobe Lightroom 4.2:
FX8350: 294.5 sec.
i7-3770k: 259.1 sec (+13%)
i5-3570k: 272.8 sec

Adobe Premiere ProCS6:
FX8350: 602
i7-3770k: 624 (+3.6%)
i5-3570k: 520

x264 encoding (1st pass):
FX8350: 62.07 fps
i7-3770k: 74.16 fps (+19.4%)
i5-3570k: 55.23 fps

x264 encoding (2nd pass):
FX8350: 15.16 fps (+4.1%)
i7-3770k: 14.55 fps
i5-3570k: 12.1 fps

3ds max 2011(CPU):
FX8350: 8.28
i7-3770k: 9.76 (+17.8%)
i5-3570k: 8.78

3ds max 2011 (render):
FX8350: 6.34
i7-3770k: 6.85 (+8.0%)
i5-3570k: 6.01

http://www.anandtech.com/bench/Product/697?vs=551

POV_RAY 3.7RC6 Multithreaded:
FX8350: 1504.4 (+10.3%)
i7-3770k: 1363.6

POV_RAY 3.7RC6 Singlethreaded:
FX8350: 252.1
i7-3770k: 312.4 (+23.9%)

Cinebench R10 Singlethread:
FX8350: 1.1
i7-3770k: 1.66 (+50.9%)

Cinebench 11.5 multithread:
FX8350: 6.89
i7-3770k: 7.61 (+10.1%)

POV-RAY 3.7 Beta 23 SMP:
FX8350: 5008
i7-3770k: 5408 (+7.9%)

Microsoft Excel 2007 SP1 Monte Carlo:
FX8350: 12.6
i7-3770k: 10.0 (+26%)

7-zip:
FX8350: 23407 (+2.6%)
i7-3770k: 22810

Doesn't look so drastically offset anymore does it? Hmm...that's interesting...so the FX8350 keeps within 10-20% of the i7-3770k in nearly everything but extremely singlethreaded applications with very specific uses? It also destroys the i5-3570k at nearly everything...how odd...but I thought...the i5-3570k was supposed to be the second coming or something...funny, huh?

That's so superior.

It’s funny how I get alternatively accused of being an AMD fanboy, and then an Intel on now. For the past 3 months, I’ve been called an AMD fanboy for aggressively supporting an FX 6300 over an i3, and an FX 8350 over an i5-3470.

In the end, this ratio debate means nothing in the real world, where things like higher clock speeds, more cores and relatively powerful iGPUs help AMD make up for it. Nobody sits and encodes AAC on his/her system for the whole day AFAIK!! But in the end, fact remains that AMD is still behind in IPG, by a more than significant margin.

Atleast you have enough sense to see value...though your judgement is clouded...(see above)...I never called you a fanboy either, I said you whipped the fanboys up into a feverpitch over your "sensational" supremacy numbers.

 

[8350rocks]

Die space most likely. Those probably though that it would work out and didn't predict software being as diverse and generic as it is. Basically law of unintended consequences being what it is. If their die shrinking then that frees up some space and they'll be using it to create another scheduler and a larger cache buffer / ect.

Direct comparison between PD and IB is kinda hard to do. IB has 3 ALU's per "core" while the PD only has 2 ALU's per "core" so immediately we're talking 33% less theoretical maximum integer performance per core. The FPU's capacity are comparable per core though the Intel's units are far more efficient due to their multiple pipeline design. When we're talking per chip your looking at 16 total ALU's for the eight "core" AMD PD chip and 12 total ALU's for the four "core" IB chip. So 33% more total theoretical integer performance for the AMD chip and about double the theoretical FPU performance though you won't ever see it hit maximum.

So for raw processing capacity the 8350 beats out any four core intel can produce while the Intel six core beats out the AMD part. Now actual realized capacity is a different matter entirely, BD/PD's relative newness combined with AMD's significantly lower R&D budget creates a handicap for the design that must be overcome. Intel's design is far more mature and developed, and we can see that in many benchmarks.

SR's architecture adds one more ALU to each PD Core...which is partly where the increase in IPC comes from on the hardware side. The other end of that is that the FPUs in SR will be able to run 3 FLOPS at once instead of 2, and with 2 decoders and shortened pipelines, and better branch prediction, it's supposed to be akin to going from a V6 camaro to a Corvette.

 

[de5_Roy]

Spoiler

The growth in speed increases are outpacing Intels.
Its been shown here that AMDs "IPC" if you will, isnt as bad as some thought, and compared to BD, PD showed huge gains.
SR hints ad another set of nice gains, again out distancing Intels growth, and lets just look at the first APU iterations, where no clock speeds, no tweaks, using the old VLIW approach, it was miles ahead, and being held back by BW, where the competition wasnt even fast enough to take advantage of the BW it had.
Upping it to GDDR5, with its BW, using GCN 1.1 if you will, improving clocks and power, then improving power management where AMD has been further behind Intel, we again will start seeing faster growth here as well, allowing for a more sustainable turbo, which Intel already enjoys, and is why unless these chips are locked in at a given frequency, so far, this advantage has been huge for Intel, where the growth is less from them, as we see HSW adding a better gfx solution, and in house VRMs, but the TDP is up in high usage, and the "IPC", if you will, is very flat.
So, AMDs gfx solutions, huge upside.
CPU, nice upside.
Power management, huge upside there as well, and until we see the full growth of the BD design, which we wont until EX, where perf is supposed to skyrocket, I cant see why people wont be picking up these chips, and future ones as well, which is alot more than can be said before RR, where they let BD out before it was truly ready, and why we see the current downtrend, but they arent PDs, or SRs, and these are right around the corner, in varying iterations

It's pretty obvious that you're an AMD fanboy, but no worries, as I am actually a bit biased towards AMD myself. But to blindly close your eyes and not see the truth is wrong.

Your 15-20% faster is more like the lower limit, not the upper one!! As i've just shown with 2 calculations, in which Intel is 37% and 50% faster. Its a reproducible calculation with most single threaded tests, and you can calculate them yourself when you have time.

There's nothing wrong with having subpar single core performance, especially when you can more than make up for it in multi threaded workloads like AMD does.
For example, encoding a song is only a matter of seconds, and is irrelevant, but when re-encoding videos with x264 for example, AMD usually provides much more performance per $, and that I like. The i3 vs FX 6300 is an outshining example of this

The issues lie in your "IPC" calculations...you cannot deduce IPC from anything anyone has posted...it's not feasibly possible. The limits for IPC are hard coded into the architecture because they are hardware limited. None of the numbers are anywhere near what some of you are talking about either. IPC over 40 is a bit crazy...yes, it is dictated by program's coding, but they never exceed the engineered maximum...that's why overclocking has become so popular...you can't increase IPC, but you can increase the clock cycles per second.

IPC has little relevance from software...the term is thrown about far too often. IPC is an architecture term from engineering as the maximum theoretical capabilities of the hardware itself. Somehow or another, it got attached to software, and it's actually nothing to do with software.

Coding efficiency and optimization of architecture are the advantages Intel has over AMD...those are the only advantages...

If you look at them on paper...the AMD is easily a LOT more CPU...

The issue is the hardware is not being advancely optimized by the coders...once that occurs...look out.

Intel also has better power consumption and those two advantages are HUGE advantages.

$3-5 cheaper to run per year is not a HUGE advantage...that's just a myth. You would literally have to run the i7-3770k for 20+ years to make up the initial cost difference in electric bill savings.

that's a misconception. power costs are different in different places. just 'cuz it costs less where you live does not mean it's as cheap everywhere else.
the bigger advantage of lower power consumption isn't from power bill, it's from higher perf/watt, less overall heat dissipation and better overall thermal efficiency, which is sometimes worth the extra price. i've seen c.a.l.f. try to dismiss it many times - just shows their ignorance and denial.
inb4 a bunch of c.a..l.f. start spamming this thread charting their 'negligible' power costs, let me show one from amd's side - between two amd trinity and intel core i5 laptops(no dgpu), the apu has significantly better perf/power in gaming. period. in my defense, i was gonna post desktop numbers and went to re-check tom's trinity articles... and ....yuck. so i am cherry picking one of the mobile apus that i've always liked, in order to champion one from amd.
http://www.tomshardware.com/reviews/a10-4600m-trinity-piledriver,3202-17.html

The only way how power consumption would be a total advantage is in mobile computing. People are total idiots when they say the FX 8350 has bad power consumption. Keep in mind people, It's an 8 core CPU. Intel hexicore CPUs have 130w requirement for boards correct? Imo, I think the FX 8350 is pretty decent due to how many cores it has.

And there is the thought that having less power draw means you can buy a cheaper PSU... but most people that buy a "High end" cpu can afford a big powersupply.

@Cazalan, So, is it a possibility that AMD has too much L2?

8350 doesn't just have bad power consumption, it has very bad power consumption (dual and tri-modules are far worse). since you're comparing sb-e, core i7 3930k has integrated pcie gen 3.0 controller, quad channel imc etc while fx8350 has ... what, just dual channel imc and... .... with fx, you get less performance (and too much heat) for hardware, compared to sb-e (one of the reasons intel charges for those so much). fx's only advantage is cheap price, for good reasons. i'd say it's appropriately priced (in the u.s. at least) unlike bd.
i see potential fx customers wanting to buy cheap psus all the time. they want to use overclocked fx(4.5ghz) with cards like radeon 7870xt with cheaper psus (e.g. which are not corsair hx series) like cx600 class. :lol:

 

[mayankleoboy1]

Spoiler

Ok, let's compare a wide range of benchmarks, shall we:

http://www.xbitlabs.com/articles/cpu/display/fx-8350-8320-6300-4300_7.html

WinRAR 4.2:
FX8350: 68 seconds
i7-3770k: 57 seconds (+19%)
i5-3570k: 83 seconds

TrueCrypt 7.1a
FX 8350: 259 MB/s (+13%)
i7-3770k: 229 MB/s
i5-3570k: 171 MB/s

iTunes:
FX8350: 185.1 seconds
i7-3770k: 126.8 sec (+49%)
i5-3570k: 130.2 sec

Adobe Lightroom 4.2:
FX8350: 294.5 sec.
i7-3770k: 259.1 sec (+13%)
i5-3570k: 272.8 sec

Adobe Premiere ProCS6:
FX8350: 602
i7-3770k: 624 (+3.6%)
i5-3570k: 520

x264 encoding (1st pass):
FX8350: 62.07 fps
i7-3770k: 74.16 fps (+19.4%)
i5-3570k: 55.23 fps

x264 encoding (2nd pass):
FX8350: 15.16 fps (+4.1%)
i7-3770k: 14.55 fps
i5-3570k: 12.1 fps

3ds max 2011(CPU):
FX8350: 8.28
i7-3770k: 9.76 (+17.8%)
i5-3570k: 8.78

3ds max 2011 (render):
FX8350: 6.34
i7-3770k: 6.85 (+8.0%)
i5-3570k: 6.01

http://www.anandtech.com/bench/Product/697?vs=551

POV_RAY 3.7RC6 Multithreaded:
FX8350: 1504.4 (+10.3%)
i7-3770k: 1363.6

POV_RAY 3.7RC6 Singlethreaded:
FX8350: 252.1
i7-3770k: 312.4 (+23.9%)

Cinebench R10 Singlethread:
FX8350: 1.1
i7-3770k: 1.66 (+50.9%)

Cinebench 11.5 multithread:
FX8350: 6.89
i7-3770k: 7.61 (+10.1%)

POV-RAY 3.7 Beta 23 SMP:
FX8350: 5008
i7-3770k: 5408 (+7.9%)

Microsoft Excel 2007 SP1 Monte Carlo:
FX8350: 12.6
i7-3770k: 10.0 (+26%)

7-zip:
FX8350: 23407 (+2.6%)
i7-3770k: 22810

Doesn't look so drastically offset anymore does it? Hmm...that's interesting...so the FX8350 keeps within 10-20% of the i7-3770k in nearly everything but extremely singlethreaded applications with very specific uses? It also destroys the i5-3570k at nearly everything...how odd...but I thought...the i5-3570k was supposed to be the second coming or something...funny, huh?

I would say that as of today, single threaded performance is more important than highly multithreaded perf, for majority of the general people.


EDIT : I had a big text heavy PDF. I opened it with Adobe reader11, and searched for a string. I was surprised, and depressed to see that the searching used only a single core. The search took about 10 seconds.

 

[sarinaide]

The CX600M might power a 8350 and 7870XT's, they improved efficiency over the old CX's by a considerable margin. I am using a CX430M in one of the systems, the CX430 had around 78% average efficiency while the CX430M has 84%, more 12v efficiency too so the new CX series is not bad at all.

AMD's power consumption is indicative of its process technology, since AMD/GF do not have the process capable of matching Intel its hard to just say that they must rub a few magic beans and boom low power, that said AMD have improved power efficiency or perf/watt per arch so that in a way is better than just throwing a low wattage chip with poor performance.

 

[sarinaide]

AMD unvails Innovative New APU,s, SoC's...

2013 APU Line-up Delivers a Huge Leap Forward in Performance-per-watt and Elite Software Experiences

“Temash” is AMD’s elite low-power mobility processor for Windows 8 tablets and hybrids. AMD expects “Temash” to be the highest-performance SoC for tablets in the market, with 100 percent more graphics processing performance2 than its predecessor (codenamed “Hondo.”)
“Kabini” targets ultrathin notebooks with exceptional battery life and offers impressive levels of performance in both dual- and quad-core options. “Kabini” is expected to deliver an increase of more than 50 percent in performance3 over the previous generation of AMD essential computing APUs (codenamed “Brazos 2.0.”)
“Richland” brings increases in both CPU4 and graphics along with improvements in battery life5. “Richland” APUs are expected to come bundled with a wealth of elite software experiences which include gesture- and facial-recognition, wireless connectivity directly to TVs and monitors, and prioritization of system resources when streaming video.

Taken direct for AMD themselves, the Modus Operandi for Richland;

1) Significant performance/watt improvements.
2) x86 improvements.
3) 20-40% iGPU peformance gains.

The line up also represents AMD making a distinctive effort to be competitive in the Tablet/Hybrid and Ultrabook market which has by and large been vacated due to the lack of performance offered by Brazos and Hondo respectively.

Testing and projections develop by AMD Performance Labs. The score for the 2012 AMD A10-4600M on the PCMark 7 Overall benchmark was 1965 and the 2012 AMD A8-4555M was 1650, while the “Richland” 2013 AMD A10-5750M was 2175 and the 2013 AMD A8-5545M was 1850. PC configs based on the “Pumori” reference design with the 2012 AMD A10-4600M with Radeon™ HD 7660G graphics, the 2012 AMD A8-4555M with AMD Radeon™ HD 7600G graphics, the 2013 AMD A10-5750M with AMD Radeon™ HD 8650G graphics and the 2013 AMD A8-5545M with AMD Radeon™ 8510G Graphics. All configs used 4G DDR3-1600 (Dual Channel) Memory and Windows 7 Home Premium 64-bit. RIN-4

RELATED LINKS:

A10 6700 exposed

Richland A-Series Mobiles

Biostar updated Richland compatibility with FM2 socket

Having spoken to a person that has ES ready Richland desktop and Mobility parts, asking what the general opinion is as he will not break NDA policy, there was a distictive thumbs up with the word "impressive". These are guys that were sceptical on Llano, hesitent but grew to appreciate Trintiy and now of the opinion that AMD APU's are starting to etch markets on superior performance* at a affordable cost.

*most metrics are still GPU driven, superior GPU performance and architecture leverages better overall results even though nobody will really argue that the x86 is industry leading.

Coolaler.com benches A10 5750M over A10 4600M

News on Kaveri:

Features dynamic L3 cache, for 8350 and Calazan. This is not that surprising as AMD pushes its HSA the sacrifice of L3 is temporary but in some instances ie: SC 2 the absence of L3 hurts the A10 by over 10FPS relative to a 965BE and 4300, in BF3 the absence of L3 is less pronounced and the A10 holds pace with the above SKU's. Since Excavator will likely be a unified socket sacrificing L3 long term would impare performance. It is just a matter of AMD evolving its IMC which SR seems to be heavily focused on.

 

[$hawn]

Doesn't look so drastically offset anymore does it? Hmm...that's interesting...so the FX8350 keeps within 10-20% of the i7-3770k in nearly everything but extremely singlethreaded applications with very specific uses? It also destroys the i5-3570k at nearly everything...how odd...but I thought...the i5-3570k was supposed to be the second coming or something...funny, huh?

That's so superior.

Don't pretend to be naive, and try to change the topic. The argument was with respect to single threaded IPG ratios from the very start of your discussion with me.
You cannot escape by throwing in multi-threaded tests in here. We are not fools here in this forum. It's a long established fact that AMD does very good in stuff that can use all it's cores. Well threaded applications are AMD's only saving grace, apart from graphics, and even there, Intel is 5% to 10% sometimes. Single core performance and excessive power draw are AMD's weak points.

As for single core IPG ratios, taking the average from the data you have posted above itself,
Avg= (1.49+1.239+1.509)/3 = 1.41, or a massive 41% faster.

In other words, its 1/1.41 = 0.70. Hence, just like I said before, this 70% performance fits comfortably inside my 65-75% window.

Btw, I'm also eagerly awaiting benchmark proof for your "1 AMD module = 2.2 Intel cores" concept.

PS:- I'm not sure if I've missed out any more single threaded tests, please do point them out if there are, so I can update my calculations.
Also, there's nothing wrong in being wrong sometimes, we're all human. Accept your fault, and learn something from it.

 

[$hawn]

I'd even say a module is bigger than 2.2 times a single core from Intel before i say its 2.2 times as good as 2 Intel cores that's just amazing maybe from some synthetic benchmark. I think its safe to say Piledrivers design is around 33% slower than Ivy's design per core per clock.

Hahaha, that's funny, but true