AMD CPU speculation... and expert conjecture

[sarinaide]

Well plenty leaks that the flagship doubles the HD7970's resources, but more importantly the GCN iteration on Hawaii seems far more revolutionary than that on GCN based HD7000 series so not only more resources but a stronger architecture as well.

The bad news will be pricing, if Nvidia hold the titan and 690 to $1000 then AMD will likely be charging more for mainstream let alone the highest end parts. I was skeptical of Nvidia first dropping its 660 and 650 generation parts to lower price points, then releasing a mere refresh. The issue is simple Nvidia have a crap load of unsold silicon that they now if these reports are true are going to have to either give away for a steal or risk not selling again. It is looking like TSMC will only have Silicon for Nvidia's new arch by 2015 so that could mean around 15months of complete market control if AMD gets its pricing right, this could result in a massive cashflow on top of the lucrative console contracts.

Kaveri which GCN cores ?

 

[The Q6660 Inside]

Being "difficult" does not mean "impossible". Difficulty =/= Impossible.

I've said on many occasions that it's not easy to design programs to scale well (scalability is the proper term for what we're describing). It requires you to go back to the drawing board and redesign the problem into one that can be attacked that way. That is ~EXACTLY~ what the Sony representative was saying, they have to go back and redesign their engine to scale well.

This isn't an option, this is as inevitable as the upgrade to 64-bit processing and the sun rising. People can kick, scream and cry all they want while viciously holding onto their dual-core CPUs, no matter as it will happen without their express consent. We are now in a period of quad-core computing, we'll be doing six / eight core computing in about three years, give or take one. We will most likely stay around 8 for five to six before needing to move on, but it's really hard to predict that far out with any sort of accuracy.

While I agree quad cores will run games better in future a =n intel hyperthreaded dual core should still provide playable fps. Lets not forget a core i3 3220 beats phenom 2 x4 and core 2 quads at games so if the dual core is faster it can still outperform quads.

Right...........................................................................................................................
965BE>3220, end of story.

http://www.tomshardware.com/reviews/ivy-bridge-wolfdale-yorkfield-comparison,3487-10.html
A core 2 quad q9550 at 3.4ghz loses on games compared to an i3 3220. Even if the 965be is faster then a q9550 at 3.4ghz it still wouldn't make the gap. I bet if you used a machine with an i3 3220 and a 965be for gaming without getting told which one was which you would think the i3 is the better cpu for gaming.

The 965BE is superior clock2clock to the Core 2 Quad, not to mention it can overclock with ease to 4.0GHz, which the 3220 is locked to 3.3. http://www.techspot.com/articles-info/642/bench/CPU_03.png Whoops, did I spoil your fun?

@sarinaide The 9970 seems to be absolutely monstrous. This reminds me of the GTX 580 situation, where the 7970 and 7950 caught nVdia in the @$$ and had market control (For a logical, informed user) for quite a bit of time.

 

[sarinaide]

[/quotemsg]http://www.tomshardware.com/reviews/ivy-bridge-wolfdale-yorkfield-comparison,3487-10.html
A core 2 quad q9550 at 3.4ghz loses on games compared to an i3 3220. Even if the 965be is faster then a q9550 at 3.4ghz it still wouldn't make the gap. I bet if you used a machine with an i3 3220 and a 965be for gaming without getting told which one was which you would think the i3 is the better cpu for gaming.

http://www.anandtech.com/bench/product/102?vs=677
http://www.cpubenchmark.net/cpu.php?cpu=AMD+Phenom+II+X4+965
http://www.cpubenchmark.net/cpu.php?cpu=Intel+Core+i3-3220+%40+3.30GHz

They are very similar cpus but games i3 wins due to better ipc and memory bandwidth.

i3 3220 better then fx8350 at games tomshardware tested, not bad for a HT dual core right:
http://www.tomshardware.com/reviews/gaming-processor-frame-rate-performance,3427-9.html[/quotemsg]

Can you change your name to herp-a-derp?

Being "difficult" does not mean "impossible". Difficulty =/= Impossible.

I've said on many occasions that it's not easy to design programs to scale well (scalability is the proper term for what we're describing). It requires you to go back to the drawing board and redesign the problem into one that can be attacked that way. That is ~EXACTLY~ what the Sony representative was saying, they have to go back and redesign their engine to scale well.

This isn't an option, this is as inevitable as the upgrade to 64-bit processing and the sun rising. People can kick, scream and cry all they want while viciously holding onto their dual-core CPUs, no matter as it will happen without their express consent. We are now in a period of quad-core computing, we'll be doing six / eight core computing in about three years, give or take one. We will most likely stay around 8 for five to six before needing to move on, but it's really hard to predict that far out with any sort of accuracy.

While I agree quad cores will run games better in future a =n intel hyperthreaded dual core should still provide playable fps. Lets not forget a core i3 3220 beats phenom 2 x4 and core 2 quads at games so if the dual core is faster it can still outperform quads.

Right...........................................................................................................................
965BE>3220, end of story.

http://www.tomshardware.com/reviews/ivy-bridge-wolfdale-yorkfield-comparison,3487-10.html
A core 2 quad q9550 at 3.4ghz loses on games compared to an i3 3220. Even if the 965be is faster then a q9550 at 3.4ghz it still wouldn't make the gap. I bet if you used a machine with an i3 3220 and a 965be for gaming without getting told which one was which you would think the i3 is the better cpu for gaming.

The 965BE is superior clock2clock to the Core 2 Quad, not to mention it can overclock with ease to 4.0GHz, which the 3220 is locked to 3.3. http://www.techspot.com/articles-info/642/bench/CPU_03.png Whoops, did I spoil your fun?

@sarinaide The 9970 seems to be absolutely monstrous. This reminds me of the GTX 580 situation, where the 7970 and 7950 caught nVdia in the @$$ and had market control (For a logical, informed user) for quite a bit of time.

HD9850 = GTX Titan

 

[The Q6660 Inside]

Being "difficult" does not mean "impossible". Difficulty =/= Impossible.

I've said on many occasions that it's not easy to design programs to scale well (scalability is the proper term for what we're describing). It requires you to go back to the drawing board and redesign the problem into one that can be attacked that way. That is ~EXACTLY~ what the Sony representative was saying, they have to go back and redesign their engine to scale well.

This isn't an option, this is as inevitable as the upgrade to 64-bit processing and the sun rising. People can kick, scream and cry all they want while viciously holding onto their dual-core CPUs, no matter as it will happen without their express consent. We are now in a period of quad-core computing, we'll be doing six / eight core computing in about three years, give or take one. We will most likely stay around 8 for five to six before needing to move on, but it's really hard to predict that far out with any sort of accuracy.

While I agree quad cores will run games better in future a =n intel hyperthreaded dual core should still provide playable fps. Lets not forget a core i3 3220 beats phenom 2 x4 and core 2 quads at games so if the dual core is faster it can still outperform quads.

Right...........................................................................................................................
965BE>3220, end of story.

http://www.tomshardware.com/reviews/ivy-bridge-wolfdale-yorkfield-comparison,3487-10.html
A core 2 quad q9550 at 3.4ghz loses on games compared to an i3 3220. Even if the 965be is faster then a q9550 at 3.4ghz it still wouldn't make the gap. I bet if you used a machine with an i3 3220 and a 965be for gaming without getting told which one was which you would think the i3 is the better cpu for gaming.

The 965BE is superior clock2clock to the Core 2 Quad, not to mention it can overclock with ease to 4.0GHz, which the 3220 is locked to 3.3. http://www.techspot.com/articles-info/642/bench/CPU_03.png Whoops, did I spoil your fun?

@sarinaide The 9970 seems to be absolutely monstrous. This reminds me of the GTX 580 situation, where the 7970 and 7950 caught nVdia in the @$$ and had market control (For a logical, informed user) for quite a bit of time.

The i3 3220 is on par if not faster then the a10-5800k multithreaded wise yet the i3 3220 loses significantly, its obvious the benchmark figures for i3 is messed up here.

crysis 2 the i3 beats the a10 cpu so crysis 3 will most likely be similar.
http://www.bit-tech.net/hardware/2012/11/26/intel-core-i3-3220-review/6
i3 3220 beating 965 phenom 2 average fps by a good margin.
http://www.tomshardware.co.uk/crysis-3-performance-benchmark-gaming,review-32644-8.html

Also core 2 quads have slightly better ipc then the phenom 2 series, everyone knows this.

Right. Mr HerpaDerp..: http://www.legionhardware.com/articles_pages/clock_for_clock_core_i5core_i7core_2_quad_and_phenom_ii_x4_performance,2.html Want a Q9550?, just add 10% to the 6600. The 3220 is not equal to the 965BE per clock, but in games that actually take advantage of even 4-cores, the easily overclocked 965BE runs away.

 

[griptwister]

I hope AMD calls it the HD 9999. LOL! What would be fantastic is if AMD Undercut the Titan by $100. (Which I highly doubt)

I'm curious to see if the HD 9870 will have 384-bit Interface. I'd probably buy one if they're cheap enough. If not, I'd imagine Nvidia would have to lower the price on their "Mini Titans." I'm looking forward to these new GPUs, even if we haven't hit 22nm in the GPU process yet! These GPUs make me feel like my soon to be 144hz monitor is a sound investment (Friend is cutting me a deal

 

[etayorius]

Being "difficult" does not mean "impossible". Difficulty =/= Impossible.

I've said on many occasions that it's not easy to design programs to scale well (scalability is the proper term for what we're describing). It requires you to go back to the drawing board and redesign the problem into one that can be attacked that way. That is ~EXACTLY~ what the Sony representative was saying, they have to go back and redesign their engine to scale well.

This isn't an option, this is as inevitable as the upgrade to 64-bit processing and the sun rising. People can kick, scream and cry all they want while viciously holding onto their dual-core CPUs, no matter as it will happen without their express consent. We are now in a period of quad-core computing, we'll be doing six / eight core computing in about three years, give or take one. We will most likely stay around 8 for five to six before needing to move on, but it's really hard to predict that far out with any sort of accuracy.

While I agree quad cores will run games better in future a =n intel hyperthreaded dual core should still provide playable fps. Lets not forget a core i3 3220 beats phenom 2 x4 and core 2 quads at games so if the dual core is faster it can still outperform quads.

Right...........................................................................................................................
965BE>3220, end of story.

http://www.tomshardware.com/reviews/ivy-bridge-wolfdale-yorkfield-comparison,3487-10.html
A core 2 quad q9550 at 3.4ghz loses on games compared to an i3 3220. Even if the 965be is faster then a q9550 at 3.4ghz it still wouldn't make the gap. I bet if you used a machine with an i3 3220 and a 965be for gaming without getting told which one was which you would think the i3 is the better cpu for gaming.

The 965BE is superior clock2clock to the Core 2 Quad, not to mention it can overclock with ease to 4.0GHz, which the 3220 is locked to 3.3. http://www.techspot.com/articles-info/642/bench/CPU_03.png Whoops, did I spoil your fun?

@sarinaide The 9970 seems to be absolutely monstrous. This reminds me of the GTX 580 situation, where the 7970 and 7950 caught nVdia in the @$$ and had market control (For a logical, informed user) for quite a bit of time.

The i3 3220 is on par if not faster then the a10-5800k multithreaded wise yet the i3 3220 loses significantly, its obvious the benchmark figures for i3 is messed up here.

crysis 2 the i3 beats the a10 cpu so crysis 3 will most likely be similar.
http://www.bit-tech.net/hardware/2012/11/26/intel-core-i3-3220-review/6
i3 3220 beating 965 phenom 2 average fps by a good margin.
http://www.tomshardware.co.uk/crysis-3-performance-benchmark-gaming,review-32644-8.html

Also core 2 quads have slightly better ipc then the phenom 2 series, everyone knows this.
Look here if you don't believe me, amd last 5 years take more electricity need to be higher clocked to get to intel levels:
http://www.anandtech.com/bench/product/49?vs=80

The Q9650 matches equally against the 955, the PhenomII 955 is clocked at 3.2 while the Q9650 is Clocked at 3Ghz, against the 965BE the Q9650 has nothing against it, even less compared to the 980BE.

There is about 200Mhz in favor of Intel "Clock for clock", but at the end of the day what matter is "Overall performance", and PhenomII was faster and cheaper that the Core2 Quads, even with the 200Mhz in favor of the Core2 Quads.

Most Intel fans are always trying to compare Clock for Clock, look at it the same way AMD competes against nVidia, AMD gives more Cores in the Radeons Compared to the GeForce Cuda Cores, the Cuda cores are indeed Faster.... but AMD gives you more cores which in turn work equally or better than the Cuda Cores, and this may be the same as the i7 Against the FX CPUs.

At the end of the day what matters is the overall performance, AMD uses faster Clock Speed, more Cores and Cheaper prices to compete Against Intel, overall the FX-8350 competes closely to the i7 3770 in Multi Threaded Apps.

You people want AMD to use the exact same Arch as Intel there for always comparing the "IPC and Clock for Clock", i seen the FX8350 give the exact same FPS as the 3770 or 4770 with the same GPU at high resolutions with Ultra or Very High Quality... the tides turn in favor of Intel massively when you start turning to lower resolutions or start lowering the Game Settings... but who the heck games lower than 1920x1080? seriously, there are TONS of benchmarks out there which puts the FX 8350 equally to the i7s at the highest resolutions/settings , people compare the FX8350 against the i5 but TekSyndicate showed how the 8350 competely destroyed the i5 in pretty much everything.

Same fps at the max settings for cheaper price? that`s awesome don`t you think? having 100 extra fps at the lowest game settings are useless, i bet most of you game on either High or Very High Settings, so those benchmarks showing massives fps at lower settings are USELESS in real world gaming.

 

[etayorius]

And there you go again comparing different CPUs when the ones you used get thrashed by AMD equivalent.

Oh by the way you quoted the wrong person, it was me not pallading9479 who posted that last reply.

Oh and don`t worry too much about Single Threading in the future, it will become irrelevant in the next 2 years, IPC will always be somewhat relevant though, but the days of Intel thrashing every game with Single Threading (Skyrim or Dirt2) are about to be gone.

We all seen the benchmarks where AMD beats Intel when the Game or Application is using true Multithreading, i would say that if a FX8350 gives a nice competition to the i7 4770 in Multithreading, Steamroller will most probably thrash the floor with the 4770... IF AMD does have plans for a Steamroller FX which i am now starting to think there will not be any FX anymore.

 

[juanrga]

lol the 8 core jaguar in the ps4 performs similar to an i3 if not slower.

Except that benchmarks show that a 4-core jaguar has about the same performance than an i3 SB.

The PS4 will likely be on par with a FX6300, 7870 and 8GB of RAM similar to what Beast said. The FX6300 isnt exactly an i3, but it isn't exactly an i7 either...

The dev. kit for the PS4 uses a 8-core FX chip, which has about the same performance than an i7 (multi-threading).

The PCs selected by Sony, Microsoft, Epic... to reproduce the performance of the new consoles include an i7 plus a GTX-680 (or superior) plus 12 (or more) GB RAM.

Some of you folks who are under-estimating Jaguar are making a massive mistake. Instructions and optimizations can make a huge difference.

I've benchmarked Gentoo FX 8350 Blender vs Windows FX 8350 Blender vs Windows 3930k Blender. FX was about half the speed of 3930k in Windows but FX actually beat 3930k when the FX was in Gentoo and the 3930k was in Windows.

3930k was at 4.2ghz and FX 8350 was at a little under 5ghz.

If you think a Jaguar with all instructions being used and an optimized OS is going to compare to an i3 running generic code for many different types of CPUs on an OS that's capable of running on a Pentium 3 (Pentium 3 only supports SSE1 and MMX) then you're in for a really bad surprise.

The question that's been on my mind is if AMD can put an APU into AM3+ and just use it as a GPGPU unit. There's no way you could actually use it as an iGPU but you could at least run OpenCL and HSA on it (maybe?).

Effectively. In my original post I made clear that I was comparing the PS4 APU to an i3/i7 on a PC. And I mentioned Windows overhead, which is of about 2x as game developers know.

I believe that the answer to your last question is negative. However, I have been thinking a while about a pair of question that are very similar to yours. The first is if a new version of dual graphics for HSA APUs could use the full iGPU for compute whereas the dGPU is used for graphics.

The second is if the iGPU of an APU HSA could be fused with a HSA enabled dGPU for dynamic split into compute or graphics according to workloads.

 

[cowboy44mag]

And there you go again comparing different CPUs when the ones you used get thrashed by AMD equivalent.

Oh by the way you quoted the wrong person, it was me not pallading9479 who posted that last reply.

Oh and don`t worry too much about Single Threading in the future, it will become irrelevant in the next 2 years, IPC will always be somewhat relevant though, but the days of Intel thrashing every game with Single Threading (Skyrim or Dirt2) are about to be gone.

We all seen the benchmarks where AMD beats Intel when the Game or Application is using true Multithreading, i would say that if a FX8350 gives a nice competition to the i7 4770 in Multithreading, Steamroller will most probably thrash the floor with the 4770... IF AMD does have plans for a Steamroller FX which i am now starting to think there will not be any FX anymore.

There's no point in trying to speak logically to hafijur, he still lives in Intel-land where he has to check with the Intel gods of electronics if something is true or not. Hence where he gets the i3 3220 dominates the FX-8350 in gaming I find it funny how die hard Intel fanboys can't admit that the days of dual core gaming are numbered. Everyone else can admit it, but they cling to this i3's will still be gaming better than top end AMD in 2017 bull crap.

I am wondering if AMD didn't postpone or cancel their Steamroller FX line when they started to do some real testing with Kaveri. If Kaveri can compete toe to toe with Intel AMD may feel that they don't need to release a Steamroller FX as it may be somewhat of a step backwards. Several months ago if someone would have speculated that Hawaii based GPUs would dominate nVidia Intel fanboys would have been all over it with how great GTX is and how bad AMD GPUs are. Now AMD is poised to dominate the high end GPU market and nVidia has nothing to compete against it, yet just a month or so ago I can remember Intel fans (hafijur if I'm not mistaken) saying they feel sorry for AMD because they can't compete against Intel in CPUs and they cant compete against nVidia in GPUs. Funny how a little bit of time makes a fool out of a fanboy spouting off.

Time will tell the story, but if Kaveri makes as big of a splash as Hawaii GPUs then AMD may feel as if they don't need Steamroller FX. While I don't expect AMD to regain the single core performance crown, they just may widen the gap in multi-core performance while narrowing the gap in single core performance. Granted it would have to be a massive gain for an APU to make (to take the place of FX and compete against i5 and i7), but its not out of the realm of impossible.

 

[juanrga]

In my previous post, I provided a quote from a game developer explaining how his multithreaded engine was not really using all the cores. This is typical for several current engines that are advertised as multithreaded, but are far from the ideal. I don't want to give names here.

I'm assuming you are referring to Sony's remakes about Planetside?

http://www.eurogamer.net/articles/2013-08-05-sony-explains-lack-of-planetside-2-pc-and-ps4-cross-platform-play-teases-character-transfers

"It really is a challenge to optimize high-end PC games to be able to work on the pantheon of hardware that's available to players nowadays, it's just insane.

"The PS4 is a much more consistent, stable platform for us to be able to develop for. The big challenge with the PS4 is its AMD chip, and it really, heavily relies on multi-threading. We have the exact same kind of Achilles heel on the PC too. People who have AMD chips have a disadvantage, because a single core on an AMD chip doesn't really have as much horsepower and they really require you to kind of spread the load out across multiple cores to be able to take full advantage of the AMD processors.

"Our engine sucks at that right now. We are multi-threaded, but the primary gameplay thread is very expensive. The biggest piece of engineering work that they're doing right now, and it's an enormous effort, is to go back through the engine and re-optimise it to be really, truly multi-threaded and break the gameplay thread up. That's a very challenging thing to do because we're doing a lot of stuff - tracking all these different players, all of their movements, all the projectiles, all the physics they're doing.

"It's very challenging to split those really closely connected pieces of functionality across in multiple threads. So it's a big engineering task for them to do, but thankfully once they do it, AMD players who've been having sub-par performance on the PC will suddenly get a massive boost - just because of being able to take the engine and re-implement it as multi-threaded.

So you're right back to the root of the problem: The main gameplay thread is hard to break up in such a way that you don't end up reducing performance due to threads locking eachother out. Its doable on the PS4/XB1, due to being able to code to the metal, and being able to guarantee when threads run, less so on the PC.

Look, I've been writing software 20+ years. I've never delivered a program that has any significant problems. I've worked on highly threaded parallel processors, and I've worked on aged 16-bit minicomputers. The assumption by many here, that you can just arbitrarily thread things out (EG: Give a separate thread to each different part of the engine) is wrong. Trying to thread in such a way reduces performance due to various stalls (thread communication, deadline management, kernel lockout, etc). It takes a LOT more time and effort, and from my experience, you'll eventually find some setup your solution has significant performance issues.

So you'll continue to see what you are seeing today: The majority of gaming benchmarks favoring Intel due to its higher IPC, with a handful going AMD's way due to AMD's high clock speed. That isn't going to be changing anytime soon. If you have two threads that result in 80% of the process load, that favors the processor with at least two fast cores. Now say we manage to spread that workload over four threads instead. That STILL favors Intel, due to 80% of the work being on four threads, thus a faster quad would still outperform a slower 8-core CPU. AMD wouldn't see a performance lead, at the same clock as Intel, until you manage to split the load evenly between 8+ threads, which is not possible for games. Period.

3 years ago, the mantra in the BD forums was "Buy BD; games will catch up in 3 years". 3 years later, the mantra is "Buy ST; games will catch up in 3 years".

You don't need to assume. I clearly said in my post that I was quoting a Planetside 2 developer.

I recall that I also said you clearly that multithreading programming was not easy. Programming a single thread in the hope that a hardware upgrade will solve your performance problems is a trivial task.

But effective multithreading programming can be done. I am not speculating. It is a reality. There exist game engines that utilize multi-core chips very well. Under those engines a 8-core FX chip is faster than a quad i5 and can even outperform an i7. This has been benchmarked. We also saw engines for future PS4 games that use efficiently six cores (exclusively for the game) without performance penalties neither bottlenecks.

The own developer that you re-quote explains how they are rewriting their old engine, transforming it in a real multithreading engine.

We know the list of PS4/XboxOne games, for all them multi-core AMD chips will see a huge performance boost because are multithreaded games.

Only bad programmers will continue to rely on IPCs and strong single cores for solving their code problems. Those programmers will not release games for the next consoles, unless they release simpler games where performance is unimportant.

I find fascinating that you claim that one cannot split the load evenly between 8+ threads, some days after Carmack gave his famous annual conference. In this occasion he mentioned how he wouldn't be surprised if next consoles were 16 ARM cores...

 

[juanrga]

Has anyone read this:

http://vr-zone.com/articles/hold-for-publication-why-its-a-big-deal-that-amd-is-delaying-kaveri/49389.html

It pretty much says the FX 9590 is the Last FX... so no more SR FX and most probably no more AM3+ CPUs, i believe only APUs for FM2+, it may just be true since we know AMD scrapped anything related to CPUs in their Roadmap...

In a previous post http://www.tomshardware.co.uk/forum/352312-28-steamroller-speculation-expert-conjecture/page-123#11287059

I explained my belief on why the FX-9000 series is the last FX series. Recent VR-zone article seems to confirm it.

Just a note: AMD continues releasing CPUs. The new Warsaw chips will be CPU only, whereas the new Berlin are coming as APU and as CPU.

 

[noob2222]

The Q6600 inside your delusion speaks volumes. A q6600 is older cpu not compareable to a q9550 for games as there is completely new instruction sets on the q9550. Yes games will run similar but there are games the 45nm quad will do far better. The i3 3220 should beat the 965be even oc to 4.2ghz in most games. I even showed you benchmarks to show i3 is faster and core 2 quad is faster ipc wise clock for clock then phenom 2 but you are delusional. AMD fanboy mentality ignore all the benchmarks, find anomaly benchmarks to suit there arguments when over 90% of the time the i3 will beat the 965be at games and the same is true for the same 45nm core 2 quad at same clock speeds to a phenom 2 x4.

Steamroller has little to no chance of catching up intel on single threaded performance so games will most likely continue the trend intels entry level core i series cpu beating amds top end cpus when gaming.

edit:
Using a 2007 or 2006 based architecture cpu in the q6600 and comparing it to a late 2009 based cpu to prove a point that a phenom 2x4 is better is laughable compared to a newer core 2 quad even with 10% difference.

Kinda like your comparing the FX 9590 to the cpu Intel is going to release 2 years from now? This entire rhetoric post is the pot calling the kettle black, your doing exactly what you claim everyone else is.

http://translate.googleusercontent.com/translate_c?act=url&depth=1&hl=en&ie=UTF8&prev=_t&rurl=translate.google.com&sl=auto&tl=en&u=http://pclab.pl/art50000-19.html&usg=ALkJrhiVTfFJ6Zp6dPlqfLUCx_NcswNZnQ

there is 14 games in here, maybe 1-2 where the I3 pulls away from the phenom x4, and 1-2 where the phenom pulls away from the I3. Most of them the I3 is right there with the phenom II, 1-2 fps difference both ways. Most of the time the phenm II 945 (3.0 ghz) is right with the q9650(3.0 ghz) with a few exceptions both directions.

The i3 3220 should beat the 965be even oc to 4.2ghz in most games.

not even close, they even did you a favor and overclocked the 3220 to 3.5 ghz vs 4.0 ghz Phenom,

http://translate.googleusercontent.com/translate_c?act=url&depth=1&hl=en&ie=UTF8&prev=_t&rurl=translate.google.com&sl=auto&tl=en&u=http://pclab.pl/art50000-53.html&usg=ALkJrhiRdFpHCYZbASYk7sr9FSRjFV6Okw

the story remains the same, couple of games favor the I3, couple favor the X4, most are tied between the 2.

pull your brain out of your back orifice and use it properly or don't speak at all.

 

[esrever]

I hope AMD calls it the HD 9999. LOL! What would be fantastic is if AMD Undercut the Titan by $100. (Which I highly doubt)

I'm curious to see if the HD 9870 will have 384-bit Interface. I'd probably buy one if they're cheap enough. If not, I'd imagine Nvidia would have to lower the price on their "Mini Titans." I'm looking forward to these new GPUs, even if we haven't hit 22nm in the GPU process yet! These GPUs make me feel like my soon to be 144hz monitor is a sound investment (Friend is cutting me a deal

doubt AMD will put a bus like that in an x800 chip. They have been using low end GDDR chips in their cards, they can move up to 7GHZ effective chips and increase bandwidth if they need it. They can easily hit 680 level performance on 256 bit bus, putting in a 384 bit bus will be too expensive. Expect the high ened to be be 384 bit. AMD launched on a pretty immature 28nm process, I expect a pretty large increase in performance with a completely new design.

If AMD does make the 9970 as fast as the titan, I expect it to be priced like the 780. Much less than titan. 28nm is mature and they should make a healthy profit from it. And AMD really can use more momentum in any market segment they can compete in.

 

[juanrga]

If Amd get's rid of the steamroller fx they are years behind intel and will NEVER catch up in the performance area, or server area for that matter since intel will have 12+ core Xeons that kick butt, why would Amd all of a sudden not care when they market themselves so heavily in the gaming field with SSD products ram and what not if this turns out to be true and i bought a 200$ board for only my phenom not only am i dropping Amd CPU's, APU's i might even drop their video cards, unless they can make a 4 core APU that has more multitasking performance then a 8 core fx and more single core performance than a fx. Also a shame to see them leave the server market as well since those are COMPLEX chips with LOTS of transistors.

Except that AMD (and analysts) predict that AMD will stole market share to Intel in the server market, thanks to the new chips.

A quad core kaveri has about the same performance than an i5 SB with older (CPU only) software and more performance than an i7 HW with HSA software. In fact kaveri A10 has 1050 GFLOP whereas i7-4770k has 848 GFLOP.

The new quad-core Berlin APUs have about the same performance than the old octo-core Opteron that are replacing in servers. This with older (CPU only) software. With HSA software the Berlin APUs will be ahead in raw performance.

AMD claims a 8x improvement in performance per watt.

And excavator comes in 2015. And recall that Excavator is not a minor update of Steamroller, unlike Ivy Bridge being a minor update over Sandy and Haswell almost no update.

 

[juanrga]

Look at Intel. 10x the resources of AMD and still failed to design a core (Haswell) that scales well from ultra-portable to high-end desktop/server. Intel focused on mobile with Haswell and has problems both at one extreme (tablets) and at the other extreme (top desktop).

That is really the problem of a "one size fits all" approach. Everything is a compromise. There are many different target markets out there, all of which benefits from a different (sometimes radically different) architecture, CPU floorplan, and platform architecture. My best guess is that you need 3-4 different architectures and at least as many platforms to reasonably address the stack from phone to big iron server. Even Intel with its massive R&D and production budgets is up against an impossible task to try to use one architecture to serve pretty much the entire stack. I don't see Haswell as a failure as it serves well as a chip in the 10-50 watt range. The problem is that Intel needs something else for the 50+ watt performance segment and something else (NOT Atom!) for the under-10 watt range. AMD has the right idea with Piledriver being a decent 35+ watt arch, Jaguar being a good 5-35 watt arch, and their ARM chips taking everything below 5 watts. They may not have the absolute best in each category but they didn't completely strike out like trying to shove Haswell into passively cooled applications.

The i7-4770k is about a 5% faster than the i7-3770k, has some performance regressions (i.e. is slower than the chip that replaces), run hotter, and OC poor.

Haswell is a laptop chip, period. Its idle power is very low and it has low power consumption in the 2-3 GHz range where a laptop chip tops out at. Laptop chips don't overclock very well and we have seen that was true for years. I wouldn't expect it to run as well as a more desktop-oriented chip such as Sandy Bridge.

Now take a look at tablets. Haswell has a too high power consumption and Intel has tried to lie about it. In case you did not notice Intel has been caught again, a pair of weeks ago, with fake power consumption figures for its Haswell tablet chips.

AMD Jaguar chip is beyond anything Intel can offer, but if AMD was to scale jaguar above 2GHz we would see similar problems to those experienced by Intel Haswell. I think that AMD dual arch. approach, Jaguar for low freq. and Steamroller for high freq., is fine.

Bingo, now you are getting it. You need multiple arches to do well in the entire product stack. I do think Intel could best Jaguar in the ULV laptop/higher-wattage embeddable segment but they would need to design a chip specifically for that rather than try to underclock the larger and more complicated Haswell. Essentially Intel needs to make a redesigned Atom but focus on a balance of performance and power consumption rather than cheapness.

The above quotes are not from calazan but from mine. I have changed the code in the above quote to reflect this. About the content of your post, well we agree on all.

 

[juanrga]

The i3 3220 is on par if not faster then the a10-5800k multithreaded wise yet the i3 3220 loses significantly, its obvious the benchmark figures for i3 is messed up here.

crysis 2 the i3 beats the a10 cpu so crysis 3 will most likely be similar.

The A10 is faster than the i3 with multithreaded benchmarks. Here a crysis2 benchmark showing how the A10 trinity is close to an i5

And a crysis 3 benchmark showing the A10 trinity > i3

And, of course, the A10 Richland is faster than the A10 trinity...

 

[noob2222]

"The roof of all evil" is not so well multithreaded as "Welcome of the jungle". Look to the six-core performing exactly as the eight-core. Why? because that section of the game is ignoring the extra cores...

More like: Its not CPU bottlenecked, so those cores don't add anything to performance.

Hence my primary point: If the CPU is getting its work done, more cores adds NOTHING to performance. Its exactly the result that you are going to see over and over and over again.

What's going on in that benchmark is EXACTLY what I've been saying since before the BD launch: Despite the fact Crysis 3 does favor Quads (see i3 vs i5 at almost the same clock), PD simply lags behind, despite its more cores. Thats you per-core performance holding back the rest of the chip.

So even though you have a level that scales to 8 cores (up to 9, according to GPUView), 4 cores is enough to get the work done. As a result, the performance difference between PD and IB is the difference between clock/IPC, not number of cores. And IB wins that by about 15% or so, which is reflected in the benchmark results.

That's also the same reason why the i3 hangs around the FX-4300: Even though its bottlenecked, its IPC is enough to still be competitive with a quad at a much higher clock. Hence my wondering about how a i3 @ 4GHz would fare...

I will say this again. AMD needs to update their memory controller. IPC IPC IPC ... IPC is not the only thing that matters.

ya, thats a 10% increase in "IPC" from running faster memory.

older article from 2010, and a bit lopsided on the speed/timing, but 15% gain in "IPC" from memory alone.

" Hence my wondering about how a i3 @ 4GHz would fare"

wish granted, sorta, you get double+ the L3 cache. Why does that matter? even a little bit can matter.

now for the 8mb l3 cache i3 3220k.

http://translate.googleusercontent.com/translate_c?act=url&depth=1&hl=en&ie=UTF8&prev=_t&rurl=translate.google.com&sl=auto&tl=en&u=http://pclab.pl/art51956-3.html&usg=ALkJrhidH8LhNxkrSxsDwW2fNupsLvKr7g

I3 @ 4.5ghz -- sometimes slower than the 3.5 ghz I5, almost always slower than 4.0 ghz I5

 

[blackkstar]

I have been pondering that AMD will kill the FX line when they release Steamroller.

The brand name has been tainted and it no longer reminds people of superior chips. At best it reminds people of a chip that's about even in multi-thread than far more expensive Intel chips and at worse a single threaded turd.

I'm kind of thinking AMD might be able to match FX 8350 performance with half as many cores.

Now if AMD stuck with FX branding, that would basically make a 4 core Steamroller only an FX 4550 or whatever.

People would be completely confused and go with FX 8350 because "higher number!" AMD knows this, it's why they released FX 9590, there is a large mass of "enthusiasts" who don't even look at benchmarks and just think bigger is better.

I wouldn't be surprised if AMD maintained their foot hold in the same market but changed names.

Some of you folks who are under-estimating Jaguar are making a massive mistake. Instructions and optimizations can make a huge difference.

I've benchmarked Gentoo FX 8350 Blender vs Windows FX 8350 Blender vs Windows 3930k Blender. FX was about half the speed of 3930k in Windows but FX actually beat 3930k when the FX was in Gentoo and the 3930k was in Windows.

3930k was at 4.2ghz and FX 8350 was at a little under 5ghz.

If you think a Jaguar with all instructions being used and an optimized OS is going to compare to an i3 running generic code for many different types of CPUs on an OS that's capable of running on a Pentium 3 (Pentium 3 only supports SSE1 and MMX) then you're in for a really bad surprise.

The question that's been on my mind is if AMD can put an APU into AM3+ and just use it as a GPGPU unit. There's no way you could actually use it as an iGPU but you could at least run OpenCL and HSA on it (maybe?).

What about being practical and thinking about all the older software that will not use jaguar like that you know every single software that's been released since x86 came out, also everyone uses windows for normal use are we going to have a big update for this on 8,7,vista?

I'm referring to the games that are being developed. PS4 will not run every single software that's been released since x86 came out. It will run highly optimized code with compilers AMD has tampered with (in a good way).

PCs will be running the old, crappy code you're speaking of. Even the games will more than likely have to have instructions disabled. Phenom 2 has no AVX, BMI, FMA, SSE4.1, etc support and that's a CPU that still goes into new gaming rigs.

Jaguar has most, if not all of those instructions.

Intel will still remain king of games like Skyrim and Shogun 2 but that's not what I'm getting at. The future is a lot more in AMD's favor.

 

[gamerk316]

I3 @ 4.5ghz -- sometimes slower than the 3.5 ghz I5, almost always slower than 4.0 ghz I5

And oftentimes faster then the FX-6300 @ 4.7. Sometimes beats the FX-8350 @ 5. Beats the FX-8350 on aggregate:

The extra cache could be improving performance a little, 2-3% or so.

Next two are average processer performance, measured against the PII X4 965 as the baseline:

http://translate.google.com/translate?act=url&depth=1&hl=en&ie=UTF8&prev=_t&rurl=translate.google.com&sl=auto&tl=en&u=http://pclab.pl/art50000-19.html

And OC'd:

http://translate.google.com/translate?act=url&depth=1&hl=en&ie=UTF8&prev=_t&rurl=translate.google.com&sl=auto&tl=en&u=http://pclab.pl/art50000-53.html

The interesting thing to note, is how the FX-8350 starts to fall back to the pack when everyone else OC's. At stock speeds, it competes against the i5-3550. With everyone OC'd, it falls being the i5-2300.

And for those who like Price/Performance (feel free to do the currency conversions):

Stock:

OC:

 

[noob2222]

I3 @ 4.5ghz -- sometimes slower than the 3.5 ghz I5, almost always slower than 4.0 ghz I5

And oftentimes faster then the FX-6300 @ 4.7. Sometimes beats the FX-8350 @ 5. Beats the FX-8350 on aggregate:

The extra cache could be improving performance a little, 2-3% or so.

aggregates are ok, I look at individual data as well, some are lopsided giving a clear advantage on the aggregate.

im guessing by the results, this game is 100% single threaded.

been a while since I tested this one, don't remember dropping below 55 fps on my 8120. could be that particular map / benchmark area they used.

 

[juanrga]

I'm kind of thinking AMD might be able to match FX 8350 performance with half as many cores.

That would imply a 200% increase in performance per core!!! Maybe Excavator achieves that, but it is unlikely.

What I find rather feasible is that a six-core Steamroller APU can match a FX-8350.

 

[Cazalan]

From rumor to strong rumor to official.

" AMD’s ‘Kaveri’ high-performance APU remains on track and will start shipping to customers in Q4 2013, with first public availability in the desktop component channel very early in Q1 2014.

‘Kaveri’ features up to four ‘Steamroller’ x86 cores, major heterogeneous computing enhancements, and a discrete-level Graphics Core Next (GCN) implementation – AMD’s first high-performance APU to offer GCN. ‘Kaveri’ will be initially offered in the FM2+ package for desktop PCs.

Mobile ‘Kaveri’ products will be available later in the first half of 2014.


Read more: http://vr-zone.com/articles/amd-confirms-kaveri-will-be-in-the-hands-of-enthusiasts-in-2014/50308.html#ixzz2bQPgU5kh
"


At least it's coming to desktop first and then to mobile.

Technically consistent with their previous statements about launching in the second half of 2013. Just not what people eager to get their hands on it interpreted it as, being available to buy in 2013.

 

[The Q6660 Inside]

The Q9650 matches equally against the 955, the PhenomII 955 is clocked at 3.2 while the Q9650 is Clocked at 3Ghz, against the 965BE the Q9650 has nothing against it, even less compared to the 980BE.

There is about 200Mhz in favor of Intel "Clock for clock", but at the end of the day what matter is "Overall performance", and PhenomII was faster and cheaper that the Core2 Quads, even with the 200Mhz in favor of the Core2 Quads.

lol, the qx9770 was released in 2007 at 3.2ghz beats the phenom 2 x4 980 at 3.7ghz 2009 year cpu.
http://www.cpubenchmark.net/cpu.php?cpu=Intel+Core2+Extreme+X9770+%40+3.20GHz
http://www.cpubenchmark.net/cpu.php?cpu=AMD+Phenom+II+X4+980

We all know how easily you could oc lets say a q9300 past the qx9770- performance levels to. The other thing is intel cpus took less electricity to then amd cpus back then to.

Also I have 8gb ddr3 ram on my q9300 rig.
This q6600 had 2x2gb ddr2 ram vs amds ddr3 ram lol.
http://www.legionhardware.com/articles_pages/clock_for_clock_core_i5core_i7core_2_quad_and_phenom_ii_x4_performance,2.html

As a person who used a QX9770 myself for a bit, I must say, 965BE>QX9770 You do realize that the QX9770 is an Extreme edition processor that is incomparable to the 965BE. The QX9770 was released in 2008 JSYK. CPUbenchmark is not really anything valid at all, just raw power in a nutshell... Nobody cared\cares about TDP on a desktop chip. The difference between AMD and Intel TDP-wise is even closer back in the day. The 9550 is a mere 10% increase over the 6600. I am no AMD fanboy, heck, I use a 920@3.8 (Your worst nightmare) and a ASUS GTX 760, with my old 4870 being reserved to drink coaster status. I like AMD and Intel for their products, they both have their issues, Intel being more corrupt than AMD, but it is all part of the chain.

 

[MU_Engineer]

The Q6600 inside your delusion speaks volumes. A q6600 is older cpu not compareable to a q9550 for games as there is completely new instruction sets on the q9550.

The Q9550 is certainly comparable to the Q9550. The only difference in instruction set between Yorkfield vs. Kentsfield is SSE 4.1. SSE 4.1 is rarely used even today.

Yes games will run similar but there are games the 45nm quad will do far better. The i3 3220 should beat the 965be even oc to 4.2ghz in most games. I even showed you benchmarks to show i3 is faster and core 2 quad is faster ipc wise clock for clock then phenom 2 but you are delusional. AMD fanboy mentality ignore all the benchmarks, find anomaly benchmarks to suit there arguments when over 90% of the time the i3 will beat the 965be at games and the same is true for the same 45nm core 2 quad at same clock speeds to a phenom 2 x4.

Two cores plus HyperThreading is NOT equal to four real cores. We have seen that time and time again with the Tom's gaming tests. The quad-core Phenom IIs do perform better than the i3s, especially when overclocked because the i3s can't be overclocked to much of a degree.

On a side note, many AMD chips are not overclocked very well when reviewed. You get a straight multiplier overclock and that's it. AMD's chips do not run the IMC/L3 at core speed (uses a different multiplier, similar to Nehalem) and get a pretty decent boost out of bumping the L3/IMC speed. However nobody ever seems to do that. Meanwhile Intel with Sandy Bridge has pegged the L3/IMC speed to the core speed so you get overclocking of the "uncore" when you increase the core multiplier. A 965BE with a 2.4-2.6 GHz L3/IMC will certainly give a Yorkfield a run for its money per clock, much more so than it did when it was at a stock 2.0 GHz L3/IMC speed.

Steamroller has little to no chance of catching up intel on single threaded performance so games will most likely continue the trend intels entry level core i series cpu beating amds top end cpus when gaming.

Single threaded performance is becoming less and less relevant. We have very clearly hit the "wall" in single threaded performance roughly 10 years ago when we stopped seeing clock speeds double for single-core CPUs every couple of years. If single threaded performance was still the end all be all, why would we have dual-core x86 CPUs let alone ones with up to 16 cores right now?

edit:
Using a 2007 or 2006 based architecture cpu in the q6600 and comparing it to a late 2009 based cpu to prove a point that a phenom 2x4 is better is laughable compared to a newer core 2 quad even with 10% difference.

The Q6600 was most popular in late 2007 through late 2009 when it sold for $200-300 rather than the almost $900 it commanded at release in early 2007. It was very popular despite being a generation or two behind current because it was priced well and performed decently against newer chips. The 12 MB L2 Q9x50 "Full Yorkfields" were much pricier and didn't perform all that much better. The Q6600 performed at least as well as the 6 MB L2 Q9x00 and much better than the abomination of the 4 MB L2 Q8xx0 units. Many people still bought Q6600s even when the first Bloomfield i7-9xxs were out. The Bloomfields were expensive, used expensive motherboards, and required expensive DDR3. It wasn't until the i5-750 Lynnfield came out on the "budget" LGA1156 socket in late 2009 and was decently faster for not that much more money that the Q6600 finally died out. So comparing the Q6600 to the early 2009 Phenom II X4s is certainly a valid comparison since both were being purchased new by quite a few people at the same time.

Also, the Phenom II is based on the 2007 Stars architecture from the first-generation Barcelona Opterons and Agena Phenoms. It just got die shrunk from 65 nm to 45 nm and had some more L3 cache added. There were no architectural or instruction changes. You might say that Phenom II had DDR3 support while Phenom did not, but that's not exactly true. AMD just didn't turn it on with the original Phenoms, nor did it do so with the very first AM2+ Phenom II X4s. DDR3 was too expensive to justify the AM3 socket at the time those chips came out and AMD enabled DDR3 usage when the price came down to something reasonable.

 

[GOM3RPLY3R]

In that "new gen of games" mini talk we're having... There's a game in Steam that caught my attention. Not because of the [strike]gender[/strike] type of game (is it gender? lol), but of the specs/requirements to be run in Alpha.

http://store.steampowered.com/app/233250/

PC System Requirements (Minimum)
-OS:Windows Vista SP2 64-bit
-Processor:32 or 64-bit Dual Core or better
-Memory:4 GB RAM
-Graphics:Shader 3.0
-DirectX®:9.0c
-Hard Drive:2 GB HD space
-Additional:These specs are still being determined.

Heh, 64bits for an alpha build... Looks like we'll start getting some new gen games sooner than later.

Cheers!

Does that happen to be Planetary Annihilation? The first day my friend showed it to me, the requirements were:

3.0 ghz Quad core or better
8 GB of RAM
5 GB of HDD

and some other really high end specs. But now its a 3.5 Ghz Dual Core or better, 4 GB of RAM, and 4 GB of HDD, and it's like a "Tycoon" game. Pretty funny.