1. This site uses cookies. By continuing to use this site, you are agreeing to our use of cookies. Learn More.

News AMD launches Piledriver-powered FX processors

Discussion in 'Article Discussion' started by Gareth Halfacree, 23 Oct 2012.

  1. fluxtatic

    fluxtatic What's a Dremel?

    Joined:
    25 Aug 2010
    Posts:
    507
    Likes Received:
    5
    As someone who got an AM3+ board shortly before BD dropped, only to be colossally disappointed, Vishera doesn't look so bad.

    Competing against IB at price points, (4300 vs i3-3220, 6300 vs i5-3470, etc) it stacks up fairly well in everything but strict single-threaded applications.

    I think it is partly that AMD tried to develop a server chip and go down from there (and apparently didn't do too badly adapting it to mobile considering its roots) and so, in a way, is ahead of its time. If the pace of software had kept up, AMD would be sitting a lot prettier. Given what a pain in the ass it apparently is to redevelop applications as multi-threaded, AMD got sort of stiffed.

    Not that I'm laying it on the doors of software developers, though - AMD should damn well have known software wouldn't be ready by the time the BD architecture was. They've been poorly managed for quite a few years now. Hopefully they let Jim Keller do his thing and stay out of his way.
     
  2. GuilleAcoustic

    GuilleAcoustic Ook ? Ook !

    Joined:
    26 Nov 2010
    Posts:
    3,277
    Likes Received:
    72
    I'm in the process to replace my ageing Q6600 w/ 4GB DDR2-800 .... either that or an IB would give a huge perf improvement. The Q6600 is not really an efficient cpu when you look at the wattage / freq / #cores / heat / perf ratio.

    Those vishera will give me a +100% perf for less than I paid for my current C2Q. I do not care to have 120 fps in games, I'm more looking at improved multi-threaded perf (rendering). I play D3 and Minecraft / Terraria .... so even an i3 would be overkill, but lacks cores when it comes to rendering.

    For people like me who play occasionnaly but like to launch sof that load all cores @ 100% simultanously, this chip is a bargain for its price, and you have to sell your organs to have something similar on Intel side. As mentioned, Intel has segmented its offering so much that it's a real nightmare when trying to have mobo + ram + cpu that have all features you need (VT-pro, ECC, etc.)
     
  3. littlepuppi

    littlepuppi Currently playing MWO and loving it

    Joined:
    26 Apr 2009
    Posts:
    3,515
    Likes Received:
    186
    Glad to see AMD have something reasonable to offer here.. Not mind blowing but not rubbish... 8350 looks to be a cracking chip for 150 quid all said and done, and if you have AM3+ you are laughing.
     
  4. The_Crapman

    The_Crapman World's worst stuntman. Lover of bit-tech

    Joined:
    5 Dec 2011
    Posts:
    7,679
    Likes Received:
    3,937
    Dear BT

    Even though you've probably finished your tests and are currently writing it all up, I have a suggestion for a test on the review to better show 'gaming performance'. Currently you only test using one top end graphics card like the GTX 590, probably so you can take the affect the GPU would have out of the equation. But the majority of us don't have that luxury, especially in cases like this where the CPU is aimed at the budget conscious.

    So, my proposed test (something I'll dub CPU-GPU Scalability) would work like this:

    -Take one gaming benchmark, e.g. BF3, 1,920 x 1,080 4xAA 16x AF, ultra detail settings, DirectX 11
    -Run that benchmark against a range of GPU's for each CPU
    -Plot a graph of the results with FPS on the Y-axis and GPU name on the X-axis

    You should end up with something like this which people could then compare the performance of the CPU against their likely GPU of choice:

    [​IMG]
    (note that these figure are complete BS that i just made up as an example)
     
    Last edited: 24 Oct 2012
  5. joco2_uk

    joco2_uk What's a Dremel?

    Joined:
    1 Sep 2009
    Posts:
    112
    Likes Received:
    2
    Does anyone know when the review will be up?
     
  6. Chicken76

    Chicken76 Minimodder

    Joined:
    10 Nov 2009
    Posts:
    952
    Likes Received:
    32
    Very good idea. I think people will be surprised to realize that with a mid-range GPU, the increase in fps from going from a Vishera CPU to an IVB i5 or i7 is going to be really small.
     
  7. Snips

    Snips I can do dat, giz a job

    Joined:
    14 Sep 2010
    Posts:
    1,940
    Likes Received:
    66
    Why change the benchmark components? So it could potentially by in favour of this product?

    AMD doesn't host the party, Bit-tech does!
     
  8. GuilleAcoustic

    GuilleAcoustic Ook ? Ook !

    Joined:
    26 Nov 2010
    Posts:
    3,277
    Likes Received:
    72
    I think it could be nice to see the CPU impact depending on GPU category (entry, mid and high end). As mentioned, not everyone can afford a GPU over 200€, so I'm not even mentioning 300, 400, and outrageously priced GPU (like the GTX590).

    Such an high end GPU needs an ultra beefy CPU .... leading to wide gap between CPUs feeding it. But with, lets say, an HD7850 or a GTX 560Ti which is more likely the case of many readers (and even that is probably too high) ... the GAP between AMD and Intel will be narrower, if not negligible.

    The question is not to favorise this CPU, but to show the difference in a REAL WORLD use (and a GTX590 is not showing what users have). Would a 300€ cpu be 50% better than a 200€ cpu if I use a mid range card ? I'm not so sure of it.

    EDIT : Sorry, but when I read a review, I'm searching for one that show the perf I'll have. I do not have and will never have a dual chip GPU ... so I do not read BT reviews. It's not representative of what I'll have at home. If I want to buy a new CPU, I read review using other components similar to the ones I have.

    and let's be honest .. ok the stock FX-8150 gave 61 fps min while a stock i7-2600k gives 92 .... but 61 fps is still more than you need .... but it is 177€ vs 314€ ... the choice is easily made for me.
     
    Last edited: 25 Oct 2012
  9. The_Crapman

    The_Crapman World's worst stuntman. Lover of bit-tech

    Joined:
    5 Dec 2011
    Posts:
    7,679
    Likes Received:
    3,937
    Not at all. It would just give a wider view of their comparative performance across the different graphics cards that people use. If you look at the results for steams hardware survey, the 560Ti is the most popular discrete graphics card followed by the 550Ti then the 460.

    The test I proposed would be of more use to the majority of people. Although yes i know those on these forums are much more likely to have a higher end card than all the users on steam, which includes my mum.
     
  10. SAimNE

    SAimNE What's a Dremel?

    Joined:
    23 Oct 2012
    Posts:
    122
    Likes Received:
    0
    actually doesnt it make sense to change the benchmark components if it results in different results???

    your view really is that because my favorite is in the lead, showing a wide range of components would be then unfair :duh:. even though it would just be a slightly wider view of the truth? if people had the time/money there would be absolutely no reason that every mid-high gpu, cpu, chipset, and ram shouldnt be tested against each other. knowing only part of the truth sucks, but unfortunately there is no real way around having just that... but at the very least people could try to show as much information as possible when reviewing products that are competing for the same range.
     
  11. Snips

    Snips I can do dat, giz a job

    Joined:
    14 Sep 2010
    Posts:
    1,940
    Likes Received:
    66
    Bit-tech does all the work for you anyway. Not on every benchmark review but they collate the data and compile the Buyers Guide. Not every month but if a component or a number of new components have impacts on one or more builds with different budgets, they do a new one.

    I'd wait for their full review of this cpu series and then see their expert knowledge and years of experience get to the conclusion you will either want or don't want to hear.
     
  12. GuilleAcoustic

    GuilleAcoustic Ook ? Ook !

    Joined:
    26 Nov 2010
    Posts:
    3,277
    Likes Received:
    72
    It's not an high end gaming CPU so it shouldn't be tested as an high end one ! The Raspberry pi is a computer ... so following your RULES it should be tested like the others .... OMG it can't run Crisis, IT'S A TRASH !!

    Yes BT does the job for us, but rules are made to be adapted. If people were all thinking like you we would all wear the same clothes, listen to the same song at the same time, drive the same car available in a single color. If you want fixed rules, go buy apple products.

    Don't be so narrow minded. Yes CPU should be tested in extrem conditions to show their limit / potential .... but it should ALSO be tested in real life condition, because not eveyone can throw 2 / 3K f***ing quid in a computer. I'm often asked to build computer for family or friend and the price tag is more 300/400€ .... up to an absolute maximum of 800€ (whole case) for GAMING and DEVELOPPMENT rig.

    Instead of testint it with 10 games, it should be doable to use only 5 (well selected) and swap for a mid range card. Same amount of time spent testing, but morei nteresting results.

    I'm more than opened to dialogue and others idea ... but I trully hate closed and narrow minded peoples. That said, I think my time in this forum as come to an end ... no dialogue possible nothing interesting coming out from threads. Bye all
     
    Last edited: 26 Oct 2012
  13. The_Crapman

    The_Crapman World's worst stuntman. Lover of bit-tech

    Joined:
    5 Dec 2011
    Posts:
    7,679
    Likes Received:
    3,937
    Maybe this will satisfy your hunger
    [​IMG]
     
  14. Shirty

    Shirty W*nker! Super Moderator

    Joined:
    18 Apr 1982
    Posts:
    12,937
    Likes Received:
    2,058
    How soon is soon?
     

Share This Page