Projected Performance: Can AMD Catch up with Intel?

I keep going back to this slide because it's incredibly relevant. It's the only indication we have from AMD of what its future roadmap will look like in terms of performance increases:

Each year AMD promised to increase performance of its high-end cores by roughly 10 - 15%. Astute observers will note that, at this rate, AMD will almost never catch up to Intel. AMD at the time was careful to point out that it's talking about 10 - 15% gains in core performance, and it could potentially see even larger increases in total chip performance by pulling other levers. Vishera is an example of AMD doing just that. The Piledriver cores by themselves don't increase performance tremendously, but they do give AMD a little more thermal headroom to work with thanks to some more efficient design decisions and better transistor choice. With Vishera, AMD took the additional power headroom and turned into a frequency advantage. The result is AMD's FX-8350 can operate in the same power envelope as the outgoing FX-8150, yet runs at an 11% higher base clock (turbo frequency remains the same). Through frequency and core level improvements, AMD was able to deliver a bit more than the 10 - 15% performance increased in promised.

If AMD is able to repeat these improvements again next year, I wondered whether or not it would get any closer to closing the gap with Intel - particularly when it came to single threaded performance. We already know from our Haswell investigations that Intel is expecting around a 5 - 15% increase in CPU performance from Haswell over Ivy Bridge. If we assume that Haswell delivers towards the 15% end of that spectrum, and if we assume that Steamroller delivers the same level of improvements that we saw from Piledriver/Vishera, we end up with some pretty interesting predictions for where things end up next year. I modeled the 2013 performance of high-end AMD and Intel platforms based on those two factors and plotted the curves in a few different benchmarks. For each generation I used the parts that AMD stacked up against one another (they are also fairly similarly priced). For 2011 I used the FX-8150 vs. Intel's Core i5 2500 and for 2012 I used the FX-8350 vs. Intel's Core i5 3570. The 2013 data is of course projected based on a 15% increase in performance from Haswell, and a repeat of the Vishera vs. Zambezi increase for AMD. This is mostly an interesting experiment so don't get too invested in the data.

We'll start with Cinebench, by far the most painful of the tests for AMD from a single-threaded performance perspective:

The Vishera gains here were decent but not enough to dramatically shrink the performance gap. Furthermore, Intel put a good amount of distance in place with Ivy Bridge and if it can continue that with Haswell I don't see much hope here.

The multithreaded Cinebench results begin in AMD's favor and remain so even with our projected performance data.

Mozilla's Kraken benchmark is another example of single threaded performance gone awry for AMD.

Thankfully, Vishera does close the gap by a decent amount and if AMD extends those gains it is on an intercept course with Intel. The bad news is, that intercept wouldn't be in 2013.

POV-Ray provides another point of view on single threaded performance, here the situation looks far less dire than under Cinebench:

Unfortunately the curves remain fairly distinct.

Once again, when we increase thread count we see AMD pull ahead.

SYSMark is a particularly telling benchmark as it is lightly threaded and does a good job of simulating all types of workloads:

The result here is AMD closing in, albeit slowly, on Intel's performance advantage. I suspect this is quite possibly the best case scenario for AMD, it doesn't necessarily want to surpass Intel in performance but it wants to get close enough where pricing and other factors (e.g. GPU performance in its APU parts) can make a bigger difference.

Our Visual Studio 2012 test is a good combination of single threaded and multithreaded workloads in one:

With Vishera, AMD did a lot to close the gap betwen itself and Intel. Another increase like this and we won't see AMD surpass Intel, but the two should remain fairly close.

These last two tests show us the other side of the coin. If both AMD and Intel continue on their present tracks, what will happen in a test where AMD already does well today?

In areas where AMD holds a significant advantage, Haswell would need to deliver more than a 15% gain in performance at the same price point to catch up.

None of the results here are all that surprising. AMD remaining on its current course isn't enough to dramatically change its standings vs. Intel in another year. Vishera definitely cut into the performance delta, but the 2013 follow-up will have to do even more to really have an impact. Steamroller is far more focused on increasing IPC, however without a new process node it'll be difficult to demonstrate another gain in frequency like we see today with Vishera. I suspect the real chance for AMD to approach parity in many of these workloads will be with its 20nm architecture, perhaps based on Excavator in 2014.

Power Consumption Overclocking
POST A COMMENT

241 Comments

View All Comments

  • CeriseCogburn - Saturday, February 02, 2013 - link


    If you came on this site to read the in depth review then do so, you idiot.

    Oh wait, instead of doing what you demand everyone else do, you go down as low as possible while holding yourself up to the angelic light and claim everyone needs to stop fanboying .... well guess what moron - if you came to read the article read it, THEN LEAVE, or stop whining in comments too, being a fanboy of sorts, the one who, doesn't FREAKING REALIZE the article itself is a BIG FAT FIGHT BETWEEN INTEL AND AMD, YOU FREAKING IDIOT.

    Have a nice day being a stupid lying fake angel.
    Reply
  • Ukdude21 - Thursday, August 15, 2013 - link

    Don't listen to him, he is just a silly bitch lol. Reply
  • nissangtr786 - Thursday, January 17, 2013 - link

    Over 2x performance per watt skower then intel and 3x slower performance per watt when both overclocked. Its crazy how far intel are ahead and the most worrying thing is haswell is where intel will be going all out to bring better performance per watt and also beat amd trump card as in integrated graphics haswell will be amazing.

    Yes amd cost less then for most of there cpu's but you pay for hat you get. Amd are releasing cpu's that intel had 3-4 years ago that were better performance per watt. Also you are saving over 2x the electricity and completing tasks much faster on intel cpu's.

    Put it this way if amd get to where intel will be with haswell in 2013, I mean for amd by 2016 or 2017 to get to where haswell will be in 2013 amd would have done a miracle as they are that far behind. I reckon amd are so far behind now that they will just target lower end market with there apu's for gaming.
    Reply
  • lordxyx - Tuesday, April 02, 2013 - link

    i just bought an FX8350 with a new mainboard and 7980 radeon. i think for an 8 core chip at 4ghz its a good price. hope it out performs my old core 2 q6600 @ 3ghz anyhow or ill feel like a complete sucker! Reply
  • ToastedJellyBowl - Friday, April 05, 2013 - link

    A lot of people posting comments in this article are nothing but tools. For those of you who can't see beyond benchmarks and who think a slight advantage in a benchmark = blowing the competition out of the water, let me give you a lesson.

    There's a difference in say a 30% difference at say 30 FPS, and a 30% difference at say, 110 FPS. When both chips are performing at 60 FPS, there is no blowing the other chip aware. At that point, it's simply a stalemate. It's just a shame that Intel fanboys are too arrogant and also ignorant to admit this. They're so fixated on "OMFG, my chip gets an extra 11.71 FPS on a benchmark than your chip does".

    Everything AMD has out there this side of a Phenom II X4 (or hell, even a low end AMD FX 4100) will run anything on the market maxed out at a solid 60+ FPS, given you are supporting it with a video card that doesn't hold it back. With that said, most people play with v-sync enabled anyways due to massive screen tearing with most games. What does it matter that a Core i7 is pulling 147 FPS and a AMD FX 8350 is pulling 107 FPS when your frame rate is just going to be locked down to 60 FPS anyways?

    I know a lot of people like to be future proofed, and the more overhead you have over 60 FPS the more future proof your system is, but future proof by an extra year != blowing the competition out of the water. Gaming requirements has pretty much hit a brick wall. System requirements has not really went up much at all in the last 2-3 years. With a Phenom II X4 965 and a GeForce 650 Ti my system runs anything I throw at it at a solid 55-60 FPS on Ultra settings. If I threw a 650 Ti Boost, or even better a 660 Ti or a 680 in my system, everything would run even better. My CPU still never really gets maxed out in most games.

    Anymore where the difference lies is how fast the CPU can encode and how fast the CPU can do other things that are not gaming related. That's where Intel is focusing right now, but as far as gaming, we've hit a brick wall, and have been behind that brick wall for several years now.

    With that being said, I'm very proud of my AMD Phenom II X4 965 that is coupled with my GeForce 650 Ti. In many games I play with my friend, this hardware compared to his Core i7-920 overclocked to over 4.0GHz running GeForce GTX 470's in SLI. In some games, I was slightly below his performance. In other games I was equal to, and in a few games my system actually outperformed his. He has since upgraded his GeForce GTX 470's with a single GeForce GTX 680, and even against that card, my system does very well in comparison. In DIRT Showdoown, we both were over the 50 FPS average mark. I was at about 57, he was at about 70, on average. Now, that may sound like a lot, right?

    Well, then you factor in the pricing. My motherboard, processor and RAM was less than $250. His motherboard alone was more expensive than everything I paid combined. Coupled with another $250 for the CPU. That's $500. That's double price just for the motherboard and processor compared to what I paid for everything, outside of a PSU, case, monitor, etc in which I already had. The performance difference, however, definitely isn't double.

    I mean, you can either go pay $600+ to build a system (motherboard, CPU, RAM since most people reuse other parts such as optical drive, sound card, network card, hard drive, PSU, etc for many years), or you can pay $250 to build a system that will get slightly less performance on benchmarks, but still be future proof.

    It's your call. I don't know about other people, but I like knowing I'm getting the best bang for the buck, and while Intel definitely may offer slightly better performance in benchmarks, AMD definitely offers the best bang for the buck. How can you turn your head at a 4 module 8 thread CPU for $185 when it costs over $300 to get a decent Intel chip? They're both future proof and will run anything at over 60 FPS for years to come, so why blow the extra $100 on the CPU and an extra $100 on a motherboard? Oh, and good luck finding an Intel motherboard that compares to the AM3+ ASUS M5A97 with a UEFI BIOS for under $200.
    Reply
  • jmcb - Monday, April 22, 2013 - link

    Most ppl in the general public will be like me. I dont OC, I tried it but never got into it. I dont even game on my PC...and for what I use my PC for....stock vs stock.....Intel is where its at. Sorry. I do lots of video encoding.

    The general public see this article....they will probably think the same thing. I also look at power consumption. Again, Intel is where its at. I had my sights set on a i7 3770k for over a year. I can probably wait 2 more years...and it might still be a better buy vs AMD.
    Reply
  • IntelBias - Sunday, June 02, 2013 - link

    I never noticed this until just now, I always heard Anandtech intel bias but never noticed it until this article. They purposely set the resolutions lower knowing that Piledriver fares much better against SB/IB in 1080p and 1440p. Reply
  • Granttamaid - Tuesday, June 11, 2013 - link

    Remember Price vs performance, AMD always win against intel. Well, intel has the fastest processor in the world,.... but, do you need all the intel's potential power? i don't think so.
    some people only use their computer for internet browsing even they have intel core i7.
    Reply
  • pantong - Monday, July 08, 2013 - link

    I see most comments talking about how this card is shit.

    Sigh.

    This card works great for what I need it to do. I host 7 servers on my computer on v boxes for my gaming community. Mine craft, ts, star made, cube world, ect... I don't get paid for my services and I need it cheap. This card give performance to host a lot of people on each on each core. From a low 4 people on cube world to 45+ on team speak.

    Why would I buy Intel for these purposes other than to spend money for same performance in the scenario and flaunt my epeen.?

    Performance is not based on score or GHz. Its based on money.
    So before saying this card is shit. Why not look at multiple applications that this card can be used it.

    As far as this scenario goes. Any unused CPU is lost money.
    Reply
  • Ukdude21 - Thursday, August 15, 2013 - link

    You keep complaining about amd fanboys, but your obviously an intel fanboy... You sound completely immature with your constant us of LOL in caps.
    No need to get too excited about cpu's... They are just part of a machine that you use lol.
    Seriously you should not post on this website if you are just going to be a immature intel fanboy.
    You sound too young to be posting on this website anyways.
    "obnoxious insolence" you should re-read your post as it is completely obnoxious.
    Your attitude stinks -.-
    Reply

Log in

Don't have an account? Sign up now