With the current rate of Intel CPU performance increases, could AMD be catching up?

[vampirr]

Read:

http://www.amd.com/us/press-releases/Pages/JimKellerJoinsAMD-2012aug01.aspx

VP's don't do design. "Cheif's" don't get in the trenches. They are managers.

So he is manager, that does not mean he wont aid and help improving AMD's Bulldozer architecture further with Steamroller. He can do that and that and interfere if he is not pleased with the progress.

This is just a role and a title, because you have a chief/manager scientist that does not mean he/she wont help in the researches and tests...

 

[Fjodor2001]

You simply cherry picked a benchmark with 2 selected CPUs. Why dont you use all the other numbers? Because they destroy your manipulating picture?

How about that?

Let me do what you do:

Haswell is now 20-75% faster than IB!

There are lots of benchmarks to choose from. I wanted to include those that are representative of the overall general performance of the CPU, which is what's relevant in most cases. I picked one generic single threaded benchmark, and one general overall performance. I think that is fair.

The single threaded benchmark I mentioned is actually the same as top one in your list. It specifies:

A8-3850 at 0.89 vs A10-5800K at 1.08. Which means 1.08/0.89=1.21 => 21% improvement, i.e. same as I mentioned in my previous post.

I could have included a general multithreaded benchmark as well, fair enough. But I have already mentioned in previous posts that the MT performance increase of Trinity is not as good as the single threaded performance increase, so that should be known. Also note that A8-5800K is a 2 module / 4 Thread CPU vs 3770K being a 4C/8T, so that comparison is not complete fair either. But you being a notorious multi-core hater MT benchmarks shouldn't matter much anyway, right...

Finally, the Haswell benchmark you mentioned is something completely different since it is not any general performance benchmark. Such specific benchmarks as AES encryption and similar is not representative of the overall performance of the CPU, which I think you already know.

 

[2is]

I'm not denying AMD is in a tough position. But at the same time I'm also looking at the actual performance improvements of the products that AMD and Intel bring out. And there it seems like Intel is stagnating performance wise, while AMD has been doing better.

There is still a CPU performance gap, which likely also explains AMD's poor economic figures the last couple of years. If AMD is too far behind, they simply will not sell (or only at heavily discounted prices). But if the gap narrows, then AMD may become an option again for a quite large market segment. Especially sine a lot of people think that CPUs are already getting fast enough for their use cases.

I'll also repeat what I said in an earlier post to clarify:

"It's pretty safe to assume that Intel and AMD learns from each other's designs. They cannot steal patents straight off, but can get "heavily inspired". That cannot be done without some lag however. So let's assume AMD is 2-3 years behind Intel in the CPU design area. Based on this:

a) If Intel increases CPU performance by 25% per year, that means Intel has:
1.25^2 to 1.25^3 => 156% to 195% the performance of AMD (being at 100%).

b) If Intel increases CPU performance by 8% per year, that means Intel has:
1.08^2 to 1.08^3 => 116% to 126% the performance of AMD (being at 100%)."

So if Intel is 50-100% faster than AMD, it might not be an option for most people. But if the gap narrows to 15-40% I think it's a different story. Especially when the better iGPU and lower price of the AMD CPUs is taken into account.

In other words, Intel will have to keep a substantial CPU performance improvement rate or the competition might catch up. Looking at the current trend where Intel improves CPU performance by ~8% per year, it's not completely unlikely that it will happen.

Intel is starting out with significantly higher performance so matching a lower performance part % for % is going to be harder. Not to mention, as AMD increases performance it will be that much harder for them to keep doing so, and as has been pointed out in pretty much every single reply to you, AMD is spending less money in that area when in fact, it will require more.

If a 2600 hits 100 fps in game X and a 3770 gets 110fps in that same game a 10% gain for 10 fps.

If a 8150 hits 80 fps in game X and an 8350 gets 89fps in that same game, that's 11.25% gain for 9 fps.

If you really think AMD can maintain performance leaps generation after generation by spending LESS money, prepare yourself for disappointment.

 

[Arachnotronic]

So he is manager, that does not mean he wont aid and help improving AMD's Bulldozer architecture further with Steamroller. He can do that and that and interfere if he is not pleased with the progress.

This is just a role and a title, because you have a chief/manager scientist that does not mean he/she wont help in the researches and tests...

AMD has Jim Keller...Intel has the army of engineers that designed Nehalem, Sandy Bridge, and Haswell. And better transistors. And more money.

Jeez...Intel is soooo screwed!

 

[Arachnotronic]

Intel is starting out with significantly higher performance so matching a lower performance part % for % is going to be harder. Not to mention, as AMD increases performance it will be that much harder for them to keep doing so, and as has been pointed out in pretty much every single reply to you, AMD is spending less money in that area when in fact, it will require more.

If a 2600 hits 100 fps in game X and a 3770 gets 110fps in that same game a 10% gain for 10 fps.

If a 8150 hits 80 fps in game X and an 8350 gets 89fps in that same game, that's 11.25% gain for 9 fps.

If you really think AMD can maintain performance leaps generation after generation by spending LESS money, prepare yourself for disappointment.

Yep.

 

[ShintaiDK]

There are lots of benchmarks to choose from. I wanted to include those that are representative of the overall general performance of the CPU, which is what's relevant in most cases. I picked one generic single threaded benchmark, and one general overall performance. I think that is fair.

The single threaded benchmark I mentioned is actually the same as top one in your list. It specifies:

A8-3850 at 0.89 vs A10-5800K at 1.08. Which means 1.08/0.89=1.21 => 21% improvement, i.e. same as I mentioned in my previous post.

I could have included a general multithreaded benchmark as well, fair enough. But I have already mentioned in previous posts that the MT performance increase of Trinity is not as good as the single threaded performance increase, so that should be known. Also note that A8-5800K is a 2 module / 4 Thread CPU vs 3770K being a 4C/8T, so that comparison is not complete fair either. But you being a notorious multi-core hater MT benchmarks shouldn't matter much anyway, right...

Finally, the Haswell benchmark you mentioned is something completely different since it is not any general performance benchmark. Such specific benchmarks as AES encryption and similar is not representative of the overall performance of the CPU, which I think you already know.

Just to burst you bubble yet again since I used a trick question. There is a 3870K. And why not use that since you like to compare the top CPU of Trinity. There was a reason why people wasnt celebrating the Trinity launch.

Those 20-30% looks pretty far away.

As I said before, you are not over the 5 stages of grief yet.

 

[frozentundra123456]

Yep.

I think I mentioned it in an earlier post, but I agree totally with this position. Yes, PD was a nice improvement over BD, but bulldozer was basically a step back, or at least a sidestep, so there was plenty of room for improvement. What I dont understand is why if there were so many improvements to be made to BD, why it came out in such a terrible state. Steamroller may be another nice improvement, but it is totally unrealistic to think they can keep making 15% improvement year after year.

 

[Phynaz]

This is just a role and a title, because you have a chief/manager scientist that does not mean he/she wont help in the researches and tests...

I give up. Okay, Jim Keller will save AMD. They are going to beat Intel in performance this year. You convinced me that Jim Keller is going to do in one year what it takes thousands of Intel engineers multiple years to accomplish.

Yep he is 4,000 times more productive than an Intel engineer.

Or, it could be that he was hired for his expertise at building an ARM team (what he did at Apple) and isn't going to be involved in x86 at all.

Which scenario do YOU think is the truth?

 

[vampirr]

Phynaz, he worked on x86 and on ARM... So he can help in both if needed...

Jim Keller is an veteran compared to most Intel's engineers and we all know that Intel wants young engineers rather keep old experienced ones that they need to pay more.

http://www.pcper.com/news/Processors/AMD-Hires-Veteran-CPU-Architect-Jim-Keller

He worked on x86, ARM, Power PC, Alpha, Mips... He has more experience than most young Intel engineers.

In other news AMD grew 12 times in India, having 20%+ of CPU market there:
http://www.lightreading.in/document.asp?doc_id=228568

 

[jpiniero]

Intel is starting out with significantly higher performance so matching a lower performance part % for % is going to be harder.

That's true, but Intel is completely focused on mobile right now. If AMD spends anything on improving the top line mainstream processors, it might be more than what Intel is spending.

 

[mrmt]

He worked on x86, ARM, Power PC, Alpha, Mips... He has more experience than most young Intel engineers.

In other news AMD grew 12 times in India, having 20%+ of CPU market there:
http://www.lightreading.in/document.asp?doc_id=228568

How many Kellers do you think Intel budget can afford?

 

[vampirr]

Intel mostly has younger engineers, I doubt they will have somebody like Jim Keller and also Intel does not treat well with their own employees... If you dint know that.

AMD can handle Intel at least in tablets, etc...

http://beta.fool.com/peterpham8/2013/03/03/amd-pounces-intel-jaguar/25663/

http://gulfnews.com/business/technology/amd-gears-up-to-take-on-intel-1.1156937

 

[2is]

Vamp, do you just make up stuff as you go or do you have a canned list of baseless comments you use?

 

[vampirr]

I don't make up that stuff at all... http://www.faceintel.com/

Intel engineers are underpayed while AMD's are well payed, source: Glass Door

 

[2is]

Guess that's why no one ever leaves AMD... Oh wait... Source: actual facts

 

[vampirr]

Guess that's why no one ever leaves AMD... Oh wait... Source: actual facts

People leave AMD because the AMD's upper managment is bad, don't know what to do since the financially AMD is in deep trouble. Its not because of salaries, people nor bad treatment to them its because of the poor moral and undecisive managment thats all.

 

[MisterMac]

I'm telling you - once\if AMD goes belly up we will miss crazies like vampir\piesquared\atenra.


I do wonder - who they will root for tho...

 

[ShintaiDK]

I'm telling you - once\if AMD goes belly up we will miss crazies like vampir\piesquared\atenra.


I do wonder - who they will root for tho...

Some already started to root for ARM.

 

[Idontcare]

I don't make up that stuff at all... http://www.faceintel.com/

Intel engineers are underpayed while AMD's are well payed, source: Glass Door

If you place zero value on job security and family stability, then yes you are underpaid at Intel and overpaid at AMD.

However, if those things represent value to you then it could just as well be the other way around.

I would not willingly subject my family to the insecurity of knowing any day I may lose my job, the kids may have to be moved and pulled away from friends and the their schools, etc, as I move about the country looking for stable work again. Others might not care about the home-life turmoil that comes with job insecurity.

Europeans probably don't like seeing their jobs go to China but I bet they like breathing their clean air and drinking their clean water. Everything has a price to it once you actually start accounting for ALL the things people value in their lives.

 

[MaxPayne63]

I'm telling you - once\if AMD goes belly up we will miss crazies like vampir\piesquared\atenra.


I do wonder - who they will root for tho...

My guess is five years from now there will still be claims that the lack of competition is going to mean $1000 quad cores... any... day... now...

 

[Idontcare]

Phynaz, he worked on x86 and on ARM... So he can help in both if needed...

Jim Keller is an veteran compared to most Intel's engineers and we all know that Intel wants young engineers rather keep old experienced ones that they need to pay more.

http://www.pcper.com/news/Processors/AMD-Hires-Veteran-CPU-Architect-Jim-Keller

He worked on x86, ARM, Power PC, Alpha, Mips... He has more experience than most young Intel engineers.

In other news AMD grew 12 times in India, having 20%+ of CPU market there:
http://www.lightreading.in/document.asp?doc_id=228568

This kind of post paints you as some kind of crazy fanatical in a cult who can't get enough of its dear leader and is just befuddled how it can be that no one else understand how holy his holiness is.

We get it. YOU think Keller is second only to your god when it comes to his powers of weaving integrated circuits with little to no resources or manpower, and on nodes that are one if not two nodes trailing the competition.

He must be one awesome dude in your dreams and the last thing I'd want to do is take away that dream.

Offhand though, if you were inclined to spend the time, you should look up Dirk Meyers and see what all he did in his life before ascending the ranks of AMD to CEO.

Notice how much he too walked on water, before he came to AMD and after, and notice what happened when his lordship moved out of the trenches and into the ranks of management where he also had the opportunity to weigh in and change things if they weren't to his pleasing.

You might notice that if Keller was truly capable of all that you wish him to be then Dirk would have already done all that and more before being fired from AMD...and yet he didn't (or couldn't).

Food for thought.

 

[Piroko]

They are investing LESS than before, meaning that they should be get LESS than what they got before. If you think they can just invest less and have. Bigger. Returns, you are expecting for a miracle, and once you go to the supernatural realm, there isn't much to discuss, is it?

You're assuming that spending more money in R&D will definitely get you a better product. Some examples which will contradict this:

Volkswagen is spending considerably less money in R&D than Toyota (3.5% of sales vs. 4.2% of sales) and yet they're growing quicker (if you exclude the one-time Tsunami hit/bump that Toyota had). And they're even more profitable - by alot.

Microsoft cut its R&D in 2005 by 25% or 1.6b$. Revenue still went up by 5b$ per year the following years.

Nokia spent massive lumps of money in R&D from 2007 onwards. Revenue still took a nosedive. In fact, until 2010 the whole Samsung group spent less in R&D than Nokia - and their Revenue is now more than 3 times as high.

I'm not expecting a miracle, shrinking R&D budget and focusing on your best products is actually one way of turning a company around without chapter 11.

 

[Vesku]

Aren't there already several "Will AMD survive?" threads this kind of discussion can go in? I haven't seen much of any true opposition to the fact that single core improvements are harder each gen and thus if AMD can keep going the gap is not likely to widen much and the probability of narrowing the gap is actually pretty good. Instead it's a lot of "well AMD won't even survive so who cares?" countered by "Nuh uh."

 

[mrmt]

You're assuming that spending more money in R&D will definitely get you a better product. Some examples which will contradict this:

Volkswagen is spending considerably less money in R&D than Toyota (3.5% of sales vs. 4.2% of sales) and yet they're growing quicker (if you exclude the one-time Tsunami hit/bump that Toyota had). And they're even more profitable - by alo1.

In R&D, there is no such thing as getting something you didn't pay, but you sometimes don't get what you paid. So unless you are assuming that amd is really inefficient in r&d, there is no chance for them to churn out chips in more complex nodes using 20% less money.

This is the conundrum that you and the others didn't provide a solution except blind faith that amd will make it.

I do think that whatever kind of company they chose to be. They won't be competing against Intel on all market bracket like before.

 

[2is]

Volkswagen is spending considerably less money in R&D than Toyota (3.5% of sales vs. 4.2% of sales) and yet they're growing quicker

A small business can spend virtually nothing in R&D and "grow quicker" than a large corporation. It's a meaningless metric when used out of context as it has been so often in this thread.