AMD CPU speculation... and expert conjecture

[FALC0N]

Where are you guys getting these numbers from? The IPC advantage Haswell has over Excavator is in the 40% to 45% range, not 60.

 

[etayorius]

Where are you guys getting these numbers from? The IPC advantage Haswell has over Excavator is in the 40% to 45% range, not 60.

I just posted some known benchmarks, and one of my own... the results seem to be larger than just 40-45% and usually 50-55% and up to 60% compared to Piledriver in Single Threaded apps and games.

If Skylake brings 20% from Haswell, is going to be epic... but i think it will be more like 10% and maybe 15% in the best case scenario, it is quite good to be honest and will be a massive jump of performance from my PhenomII at 4GHZ.

I would keep the PhenomII if DX12 and Vulkan were already heavily used in games, since it would boost my CPU performance to at least 50% making the need to upgrade irrelevant for some time, bad news is the very first Games supported by DX12 will arrive between Fall 2015 and early 2016, with Vulkan enabled games arriving possibly mid 2016 or fall of the same year, i need an upgrade no longer than October of this same year the most, this was the reason why i was so excited about MANTLE, i just wish AMD would had keept their promise of releasing 15+ games by end of 2014, that would had allowed me to keep my CPU for a bit longer, since neither DX12 and Vulkan will be heavily used in the short term, i really need to upgrade.

 

[8350rocks]

You are comparing performance for intel where it is strongest, against performance for AMD where it is weakest, and you are cherry picking benchmarks.

The level of wrongness in these claims is ridiculous.

You cannot sit back and say "AMD sucks for gaming because Skyrim..."

Look at the total picture...in multi-threaded workloads, the AMD processors are often within 15-25% of haswell.

Programs are getting more threaded, not less threaded...single core performance is relevant, not saying that it is not...however, the performance for a single core is becoming less and less the bar by which all other things are measured. Unless, of course, you are wanting to exacerbate a tech difference between 2 processor types, or, you are trying to troll a group of engineers that are all well aware that there is more to a CPU than cherry picked benchmarks on single core performance. So what if intel processor A is 60% faster at superpi? Who cares...it is x87 code...miraculously, so is skyrim. As well you are likely posting a lot of things that are similarly poorly optimized for a modern cpu architecture.

However, keep believing AMD is 60% behind...whatever...

All I can say is every AMD rig I own continues to do what I ask of it within my expectations, and within a reasonable time frame. I actually work on an intel processor at work, and frankly, if you told me the intel rig was my AMD rig at home, the biggest difference I would notice in productivity, is the fact that my PC has a SSD and boots about 3x faster.

 

[de5_Roy]

^^where did any of them mention skyrim?

 

[de5_Roy]

this deserves own post:
Mantle melded with PowerVR in real-time Vulkan demo
http://techreport.com/news/27932/mantle-melded-with-powervr-in-real-time-vulkan-demo

That session started with a fundamental truth: Vulkan is based on AMD's Mantle API, only it's been adapted and extended to work properly with a range of different modern GPU architectures.

rumors of mediatek getting radeon cores, vulkan's debut, project skybridge et al makes me speculate if amd has a deal with mediatek to build high performance mobile x86 and arm socs from amd, kinda like the deal intel has with rockchip. might financially help amd and lessen the barrier of entry to the mobile market.

 

[8350rocks]

^^where did any of them mention skyrim?

That was an example...every time someone cites single core performance in gaming, skyrim is among the benches linked...so it is a bit of a habitual response I suppose.

EDIT: Nice link too, by the way, if AMD and MediaTek team up...that could literally be another console type cash flow stream for AMD. A few more of those, and suddenly they have the R&D money to do some of the big stuff they want.

 

[esrever]

I am assuming Mediatek wants HSA and they are working with AMD to enable it. It would be much easier to partner with AMD and use radeon than to develop their own graphics IPs. It will also be good because AMD will allow mediatek to have leverage over the other ARM players who has to depend on imagination technologies in the case where AMD can make better performing mobile GPUs.

2016 might finally be the year for AMD to get back on track. Hopefully they don't mess things up.

As for piledriver vs intel, why are people looking at synthetic benchmarks running small kernel loops and cinebench when real world applications can be used? POV ray is actually a rendering software that can be used, why use cinebench which is just synthetic that tries to emulate cinima4d performance and is only ever trying to do that? And passmark has always been pointless and might as well be a random number generator, It doesn't even have a objective and is only used by sites with no real world data to get hits. You might as well be looking at sisandra numbers.

 

[etayorius]

You are comparing performance for intel where it is strongest, against performance for AMD where it is weakest, and you are cherry picking benchmarks.

The level of wrongness in these claims is ridiculous.

You cannot sit back and say "AMD sucks for gaming because Skyrim..."

Look at the total picture...in multi-threaded workloads, the AMD processors are often within 15-25% of haswell.

Programs are getting more threaded, not less threaded...single core performance is relevant, not saying that it is not...however, the performance for a single core is becoming less and less the bar by which all other things are measured. Unless, of course, you are wanting to exacerbate a tech difference between 2 processor types, or, you are trying to troll a group of engineers that are all well aware that there is more to a CPU than cherry picked benchmarks on single core performance. So what if intel processor A is 60% faster at superpi? Who cares...it is x87 code...miraculously, so is skyrim. As well you are likely posting a lot of things that are similarly poorly optimized for a modern cpu architecture.

However, keep believing AMD is 60% behind...whatever...

All I can say is every AMD rig I own continues to do what I ask of it within my expectations, and within a reasonable time frame. I actually work on an intel processor at work, and frankly, if you told me the intel rig was my AMD rig at home, the biggest difference I would notice in productivity, is the fact that my PC has a SSD and boots about 3x faster.

Well i have no idea where Skyrim comes into play, since it was never in the discussion... But i did Provided Benchmarks, now when you say 15-25% compared to Haswell you mean...?

You can do your own benchmarks, i did my own and i am seeing Intel lead in Performance Test by about 60%, AnandTech and CPUBoss support my Passmark results.

Unless there is some sort of Industry bias against AMD for unknown reasons that is making AMD look 40% slower than what it actually is? but this falls under Conspiracy Theories which has zero to do with Proof.

Now, we are comparing the top high end between AMD and Intel, i did test between i7 4790K and FX8350 and the gap is still there, i did my test against FX8350 vs the i5 4670K and the gap is still there, now i am talking about SingleThreaded IPC and i was very clear with that, since in well threaded apps AMD may not be so far behind and actually passes Intel in a few tests.

Benchmarks are all over the Internet, and i am having a hard time trying to find any benchmark that supports your claim, there is got to be something in between from what i am seeing and what you are seeing because there is clearly a discrepancy.

If we take into account Gaming, you will not really see that big of a difference since work is tends to be load more into the GPU, at higher resolutions the gap shrinks considerably to a few FPS, maybe this is what you are trying to say?

Intel performance lead is there, and is in no way just 15-25% IPC. Let´s be honest and start accepting facts... if what you are claiming was true, i don´t understand how or why are AMD CPUs are not selling that good.

I can understand your stand, you been claiming to be related to AMD one way or another, which explains somewhat your actions in these forums, sometimes you sound as if you are trying to sell AMD Products through this forum, but facts are hard to avoid when there is so much evidence all over the place. So I can`t seem to understand why you seem to be in denial.

Anyway I always been supporting AMD stuff and i don`t hate them, i think MANTLE was GENIUS and also what it achieved, and you all probably seen me comment very positive about it since i have been wanting for something like for ages, tough it really annoyed me that AMD could not keep their promise of 15+ titles before 2014 ended, the more titles the better since i would had better chances of running them better with my current CPU.

I also like nVidia products (but not their selfish and anti consumer schemes). Intel... well i never really like them because all the crap they have done to AMD and the industry, but fact is they are a better choice (By far) if you have the cash, and this time around i waited so long for AMD Promises that i now have enough cash for a premium System.

 

[etayorius]

this deserves own post:
Mantle melded with PowerVR in real-time Vulkan demo
http://techreport.com/news/27932/mantle-melded-with-powervr-in-real-time-vulkan-demo

That session started with a fundamental truth: Vulkan is based on AMD's Mantle API, only it's been adapted and extended to work properly with a range of different modern GPU architectures.

rumors of mediatek getting radeon cores, vulkan's debut, project skybridge et al makes me speculate if amd has a deal with mediatek to build high performance mobile x86 and arm socs from amd, kinda like the deal intel has with rockchip. might financially help amd and lessen the barrier of entry to the mobile market.

This is cool, thanks for sharing.

 

[Cazalan]

rumors of mediatek getting radeon cores, vulkan's debut, project skybridge et al makes me speculate if amd has a deal with mediatek to build high performance mobile x86 and arm socs from amd, kinda like the deal intel has with rockchip. might financially help amd and lessen the barrier of entry to the mobile market.

Except they would be a competitor for Skybridge. Makes little sense to me unless AMD is to abandon ARM for the consumer space. Which I suppose is possible since the margins have gone to crap.

The Rockchip/Intel thing works because they're locked into 28nm SoCs while Intel has moved to 14nm.

 

[juanrga]

You are comparing performance for intel where it is strongest, against performance for AMD where it is weakest, and you are cherry picking benchmarks.

No. The 60% is average as shown before. If you want compare performance for Intel where it is strongest, against performance for AMD where it is weakest, then chose some FP-intensive benchmark like this

http://openbenchmarking.org/embed.php?i=1406101-KH-INTELDEVI74&sha=505359a&p=2

where quad-core 4790K Haswell is 175% faster than octo-core FX-8350 because Haswell is a strong architecture with 32 FLOP per core, whereas Piledriver is a weak architecture with only 8 FLOP per core.

Look at the total picture...in multi-threaded workloads, the AMD processors are often within 15-25% of haswell.

~60% IPC per core plus the ~20% module penalty on Piledriver means ~92% more performance on integer. Thus you need 8 Piledriver cores to match the performance of 4 Haswell cores on integer workloads. Haswell quad-cores that you mention are clocked at 3.5GHz whereas FX-8350 is clocked at 4GHz. This is 14% higher clocks for AMD. And the difference between doubling cores and 92% gives another 4% for AMD.

Conclusion: quad-core Haswell at 3.5GHz is about 14--19% behind octo-core Piledriver at 4GHz in multithreaded integer code.

Programs are getting more threaded, not less threaded...single core performance is relevant, not saying that it is not...however, the performance for a single core is becoming less and less the bar by which all other things are measured. Unless, of course, you are wanting to exacerbate a tech difference between 2 processor types

Single core performance continues being fundamental for multithreaded software. Reason why Keller is investing years on improving the single core performance for Zen.

 

[griptwister]

Considering these failure AMD 8 core CPUs are pulling in the numbers pretty decently for being as out dated as they are, I'm pretty happy. I will be upgrading to Intel. Not because AMD is worse, but because Intel is currently better, and I won't spend all my time waiting on a new CPU from AMD. Same goes for graphics cards. AMD needs to get competitive quick. Hopefully this Mediatek deal buys them some time and support. I can't wait for what AMD has in the future.

 

[etayorius]

Considering these failure AMD 8 core CPUs are pulling in the numbers pretty decently for being as out dated as they are, I'm pretty happy. I will be upgrading to Intel. Not because AMD is worse, but because Intel is currently better, and I won't spend all my time waiting on a new CPU from AMD. Same goes for graphics cards. AMD needs to get competitive quick. Hopefully this Mediatek deal buys them some time and support. I can't wait for what AMD has in the future.

Totally agree.

 

[gamerk316]

1: As I said back in the BD thread, it will crush in benchmarks when all it's cores can be utilized. BD/PD is faster then Intel at full load. The problem is, the more cores, the harder to keep those cores utilized. Hence why Intel often pulls ahead.

2: As for the Mediatek thing, I see this as Mediatek trying to differentiate itself from Qualcomm. I could easily see a situation in mobile where Mantle matters until we get OGLNG/Vulkan GPU support, which could give Mediatek a HUGE performance benefit over the competition. That being said, this puts Mediatek on more or less even ground with AMD in the ARM space, since both would be making ARM cores with GCN graphic cores. So unless AMD is more desperate for money then even I think they are, this is telling me that AMD is basically abandoning ARM on mobile.

 

[de5_Roy]

EDIT: Nice link too, by the way, if AMD and MediaTek team up...that could literally be another console type cash flow stream for AMD. A few more of those, and suddenly they have the R&D money to do some of the big stuff they want.

should i read into this? amd would benefit licensing it's gpu ip to hsa members but only mediatek has no in-house gpu ip iirc. qualcomm and arm both have their own gpu ip. i'd say high end tablets, hdmi dongles and media streamers, chromebookd/android-running-ultrathins would be good places to start. can/will they sell to apple?

2016 might finally be the year for AMD to get back on track. Hopefully they don't mess things up.

i hope so. so far all the news about amd screwing up were related to marketing/management/sales stuff like the times they had excess motherboard chipsets or failed to sell to laptop oems, or had excess gpus or excess older cpus and apus.

Except they would be a competitor for Skybridge. Makes little sense to me unless AMD is to abandon ARM for the consumer space. Which I suppose is possible since the margins have gone to crap.

The Rockchip/Intel thing works because they're locked into 28nm SoCs while Intel has moved to 14nm.

i mentioned skybridge in case amd wants to build semi-custom, socket-compatible arm and x86 socs and sell them through mediatek or someone else. the arm socs will compete with mediatek's but it'd be better than nothing. i don't see amd coming to consumer space on it's own due to intel, nvidia and qualcomm's presence.
mediatek's amd-flavored socs could be good for samsung, which would gain a foundry customer as well as chip supplier (that is not qualcomm) for it's galaxy lineups while amd and mediatek gain a non-glofail foundry partner. i think it'd be a win-win...-win...? seems good for short term.

2: As for the Mediatek thing, I see this as Mediatek trying to differentiate itself from Qualcomm. I could easily see a situation in mobile where Mantle matters until we get OGLNG/Vulkan GPU support, which could give Mediatek a HUGE performance benefit over the competition. That being said, this puts Mediatek on more or less even ground with AMD in the ARM space, since both would be making ARM cores with GCN graphic cores. ■ So unless AMD is more desperate for money then even I think they are, this is telling me that ■ AMD is basically abandoning ARM on mobile.

i agree with the last two. amd will have nearly two years of no major cpu/soc launch. the past two cpu launches turned into full chip launches (sr and exc to kaveri and carrizo resp.) with the non-cpu stuff getting more publicity and attention.

as for mantle and vulkan, can they market an soc before vulkan is finalized?

 

[gamerk316]

as for mantle and vulkan, can they market an soc before vulkan is finalized?

Depends what Vulkan is targeting, HW wise. For example, Tesselation isn't required by the spec, so pretty much anything that meets OGL 3/OGL ES 2.0 should be supported by Vulkan, and pretty much all HW these days does, since those specs are aged.

My point is this: By using GCN cores, Mediatek is opening the *possibility* of using Mantle on mobile, which is where an API like Mantle would give the most benefit. Especially since power draw during gaming is a major problem on mobile, since less power leads to longer play time, which is a thing. Nevermind the performance boost.

Which again makes me wonder where AMD is in mobile. Granted, they missed the boat, but I'm continually dumbfounded they never made the attempt when smartphones took off.

 

[jdwii]

Well i don't know, Haswell is already about 60% ahead of Piledriver in IPC... if we take into account 5% more from Broadwell and 5% (minimum) more from Skylake that is a good 70% ahead of Piledriver, i honestly don't think AMD can manage a 100% more in one generation, i seriously don't buy it.

40% is good and reasonable, i would definitely buy a Zen system if it were out right now considering AMD prices seem to be cheaper most of the time compared to damn Intel, sadly 2016 is waay to late (at least for me and i have been waiting since 2011).

Every little gains in IPC helps AMD, and i really hope they manage at least 40% because anything less would just be a another fail and i doubt AMD can survive another Faildozer.

I really want to see AMD succeed, i definitely do not want to see them go or transform into VIA but the state in which their current CPUs is honestly shameful to be honest, i can understand people who need more than 2 cores using them, but as a gamer i see little need of their piledriver CPUs. Also games are barely starting to use 4 cores, not even MANTLE can take advantage of more than 4 as we seen in benchmarks with the i7 with 8 cores, i was expecting waaay more performance from that CPU, but all the gains in FPS was due to the super fast IPC instead of the 8 cores with HT.

What is everyone smoking to claim 60% lead in IPC from intel? Seriously...are you all from Amsterdam, Colorado, or Washington?

http://www.tomshardware.com/charts/cpu-charts-2013/compare,3143.html?prod%5B6234%5D=on&prod%5B5877%5D=on&prod%5B5755%5D=on

There are a few edge cases where the 4770k is about 20-25% faster than the 8350 at stock clocks for both CPUs.

However, that is not 60%, nor is it anywhere near that.

Clock for clock is irrelevant as well...because intel puts a processor out where it is comfortable being compared, and so does AMD. If you want to talk about overclocking something, compare both overclocked, or both stock. Otherwise, you are comparing apples and oranges. As it sits...we are looking at a 25% difference between haswell and FX at best.

Now, if we assume 15% for broadwell and skylake combined...

25+15 = 40%

Zen improvements = 40%

Hmm...interesting...who said they would be close when zen launched?

Wow been gone awhile(building rigs and work is picking up) 8350 rocks are you willing to do a benchmark battle to prove your point? I'll be more then willing to do it same clock speeds though!!

IPC of intel haswell is easily 45+% better per clock(on average). Has nothing to do with ego or anything else has everything to do with numbers and facts if you are willing to ignore them then fine reality won't change however.

 

[jdwii]

Where are you guys getting these numbers from? The IPC advantage Haswell has over Excavator is in the 40% to 45% range, not 60.

Yes you are correct however Piledriver isn't as nice which is what we are talking about

 

[jdwii]

You are comparing performance for intel where it is strongest, against performance for AMD where it is weakest, and you are cherry picking benchmarks.

The level of wrongness in these claims is ridiculous.

You cannot sit back and say "AMD sucks for gaming because Skyrim..."

Look at the total picture...in multi-threaded workloads, the AMD processors are often within 15-25% of haswell.

Programs are getting more threaded, not less threaded...single core performance is relevant, not saying that it is not...however, the performance for a single core is becoming less and less the bar by which all other things are measured. Unless, of course, you are wanting to exacerbate a tech difference between 2 processor types, or, you are trying to troll a group of engineers that are all well aware that there is more to a CPU than cherry picked benchmarks on single core performance. So what if intel processor A is 60% faster at superpi? Who cares...it is x87 code...miraculously, so is skyrim. As well you are likely posting a lot of things that are similarly poorly optimized for a modern cpu architecture.

However, keep believing AMD is 60% behind...whatever...

All I can say is every AMD rig I own continues to do what I ask of it within my expectations, and within a reasonable time frame. I actually work on an intel processor at work, and frankly, if you told me the intel rig was my AMD rig at home, the biggest difference I would notice in productivity, is the fact that my PC has a SSD and boots about 3x faster.

8350rocks this is coming from a long time Amd fan but i play a lot of games old and new. Dolphin emulator is a joke on even a 5Ghz 8 core fx evil within is also a joke again i ask you why buy a CPU that offers inconsistent performance? I play many games and i will say this to you when a game uses the 8350 to its fullest it competes with a I5 haswell and i7 haswell CPU but when it doesn't its worse then a I3 and in dolphin for me worse then a pentium.

 

[gamerk316]

^^ Emulators are single core by nature. Any time you need to emulate, you almost have to remain non-threaded to avoid running into various syncing problems. That's why i continue to hold the PS3 is never going to be emulated anywhere near full speed, since you'd basically need to emulate a 240 GFLOP 7 core CPU on a single core at acceptable speed.

 

[8350rocks]

You are comparing performance for intel where it is strongest, against performance for AMD where it is weakest, and you are cherry picking benchmarks.

No. The 60% is average as shown before. If you want compare performance for Intel where it is strongest, against performance for AMD where it is weakest, then chose some FP-intensive benchmark like this

http://openbenchmarking.org/embed.php?i=1406101-KH-INTELDEVI74&sha=505359a&p=2

where quad-core 4790K Haswell is 175% faster than octo-core FX-8350 because Haswell is a strong architecture with 32 FLOP per core, whereas Piledriver is a weak architecture with only 8 FLOP per core.

Look at the total picture...in multi-threaded workloads, the AMD processors are often within 15-25% of haswell.

~60% IPC per core plus the ~20% module penalty on Piledriver means ~92% more performance on integer. Thus you need 8 Piledriver cores to match the performance of 4 Haswell cores on integer workloads. Haswell quad-cores that you mention are clocked at 3.5GHz whereas FX-8350 is clocked at 4GHz. This is 14% higher clocks for AMD. And the difference between doubling cores and 92% gives another 4% for AMD.

Conclusion: quad-core Haswell at 3.5GHz is about 14--19% behind octo-core Piledriver at 4GHz in multithreaded integer code.

Programs are getting more threaded, not less threaded...single core performance is relevant, not saying that it is not...however, the performance for a single core is becoming less and less the bar by which all other things are measured. Unless, of course, you are wanting to exacerbate a tech difference between 2 processor types

Single core performance continues being fundamental for multithreaded software. Reason why Keller is investing years on improving the single core performance for Zen.

Juan, I can pull out a phoronix benchmark showing the 8350 is 500% faster than a 3930k, which by your estimations means it would be about 450% faster than skylake by extrapolation, and theoretical physics can prove an elephant will hang, just fine, from a single blade of grass on the side of a cliff.

Just because you pull out some cherry picked garbage means nothing. I trust nothing you post any way because you always post nonsensical things that are designed to exacerbate a single thing.

 

[8350rocks]

You are comparing performance for intel where it is strongest, against performance for AMD where it is weakest, and you are cherry picking benchmarks.

The level of wrongness in these claims is ridiculous.

You cannot sit back and say "AMD sucks for gaming because Skyrim..."

Look at the total picture...in multi-threaded workloads, the AMD processors are often within 15-25% of haswell.

Programs are getting more threaded, not less threaded...single core performance is relevant, not saying that it is not...however, the performance for a single core is becoming less and less the bar by which all other things are measured. Unless, of course, you are wanting to exacerbate a tech difference between 2 processor types, or, you are trying to troll a group of engineers that are all well aware that there is more to a CPU than cherry picked benchmarks on single core performance. So what if intel processor A is 60% faster at superpi? Who cares...it is x87 code...miraculously, so is skyrim. As well you are likely posting a lot of things that are similarly poorly optimized for a modern cpu architecture.

However, keep believing AMD is 60% behind...whatever...

All I can say is every AMD rig I own continues to do what I ask of it within my expectations, and within a reasonable time frame. I actually work on an intel processor at work, and frankly, if you told me the intel rig was my AMD rig at home, the biggest difference I would notice in productivity, is the fact that my PC has a SSD and boots about 3x faster.

8350rocks this is coming from a long time Amd fan but i play a lot of games old and new. Dolphin emulator is a joke on even a 5Ghz 8 core fx evil within is also a joke again i ask you why buy a CPU that offers inconsistent performance? I play many games and i will say this to you when a game uses the 8350 to its fullest it competes with a I5 haswell and i7 haswell CPU but when it doesn't its worse then a I3 and in dolphin for me worse then a pentium.

Really? It does not seem to be a problem for a 9590 juan.

 

[juanrga]

1: As I said back in the BD thread, it will crush in benchmarks when all it's cores can be utilized. BD/PD is faster then Intel at full load. The problem is, the more cores, the harder to keep those cores utilized. Hence why Intel often pulls ahead.

2: As for the Mediatek thing, I see this as Mediatek trying to differentiate itself from Qualcomm. I could easily see a situation in mobile where Mantle matters until we get OGLNG/Vulkan GPU support, which could give Mediatek a HUGE performance benefit over the competition. That being said, this puts Mediatek on more or less even ground with AMD in the ARM space, since both would be making ARM cores with GCN graphic cores. So unless AMD is more desperate for money then even I think they are, this is telling me that AMD is basically abandoning ARM on mobile.

About 2. Recall AMD did cancel its tablet plans, has not phone plans, and recently claimed is not interested in chromebooks.

 

[juanrga]

You are comparing performance for intel where it is strongest, against performance for AMD where it is weakest, and you are cherry picking benchmarks.

No. The 60% is average as shown before. If you want compare performance for Intel where it is strongest, against performance for AMD where it is weakest, then chose some FP-intensive benchmark like this

http://openbenchmarking.org/embed.php?i=1406101-KH-INTELDEVI74&sha=505359a&p=2

where quad-core 4790K Haswell is 175% faster than octo-core FX-8350 because Haswell is a strong architecture with 32 FLOP per core, whereas Piledriver is a weak architecture with only 8 FLOP per core.

Look at the total picture...in multi-threaded workloads, the AMD processors are often within 15-25% of haswell.

~60% IPC per core plus the ~20% module penalty on Piledriver means ~92% more performance on integer. Thus you need 8 Piledriver cores to match the performance of 4 Haswell cores on integer workloads. Haswell quad-cores that you mention are clocked at 3.5GHz whereas FX-8350 is clocked at 4GHz. This is 14% higher clocks for AMD. And the difference between doubling cores and 92% gives another 4% for AMD.

Conclusion: quad-core Haswell at 3.5GHz is about 14--19% behind octo-core Piledriver at 4GHz in multithreaded integer code.

Programs are getting more threaded, not less threaded...single core performance is relevant, not saying that it is not...however, the performance for a single core is becoming less and less the bar by which all other things are measured. Unless, of course, you are wanting to exacerbate a tech difference between 2 processor types

Single core performance continues being fundamental for multithreaded software. Reason why Keller is investing years on improving the single core performance for Zen.

Juan, I can pull out a phoronix benchmark showing the 8350 is 500% faster than a 3930k

First, 3930k is not Haswell, 3930k is Sandy Bridge.

Second, I invite you to give that cherry picked benchmark comparing 8350 and 3930k.

Third, phoronix did a complete review of the FX-8350 that showed that the FX-8350 is ~25% behind the i7-3770k on average. Aka

8 PD core @ 4GHz < 4 IB core @ 3.5GHz

which agrees with my above conclusion about Haswell:

Conclusion: quad-core Haswell at 3.5GHz is about 14--19% behind octo-core Piledriver at 4GHz in multithreaded integer code.

 

[logainofhades]

I suspect that they will give another try at mobile, once they have ARM at a level that it is competitive.