AMD FX-8150 Black Edition Processor Launch Review @ HH

Discussion in 'Reviews & Articles Discussion' started by craig5320, Oct 12, 2011.

Thread Status:
Not open for further replies.
  1. craig5320

    craig5320 Well-Known Member

    Joined:
    May 7, 2002
    Messages:
    9,682
    Likes Received:
    226
    Trophy Points:
    88
    AMD FX-8150 Black Edition Processor Launch Review

    Today we get to see for the first time how the top model in AMDs new FX range of CPUs, previously known as Bulldozer, performs against the existing Intel equivalent as we put the two CPUs head to head. Not only that though we will test their entire platform, testing gaming, USB 3.0, SATA 6Gb/s and PCIe performance.
     
  2. slugbug

    slugbug Going Broke Saving Money

    Joined:
    Oct 15, 2002
    Messages:
    612
    Likes Received:
    15
    Trophy Points:
    28
    Reading a few reviews and no one seems to have tested folding@home on it yet.
     
  3. Takaharu

    Takaharu Unus offa, unus iuguolo

    Joined:
    Aug 31, 2009
    Messages:
    3,350
    Likes Received:
    346
    Trophy Points:
    108
    Nice to see that AMD are happily rivalling the 2600k now, even if the CPU power does fall slightly short. I'm itching to see how the FX processors fare when paired with an Nvidia graphics card; will there be a substantial drop in performance?
     
  4. blibbax

    blibbax nahm8

    Joined:
    Sep 6, 2008
    Messages:
    6,210
    Likes Received:
    203
    Trophy Points:
    73
    I'd be very surprised if there was.

    It's odd how the AMD chip keeps up in games while falling so far behind elsewhere. I swear in some of the other tests it was further behind than the Phenom X6.

    Also, what was the power consumption like with that 5GHZ OC?

    (Nice review btw :p)

    EDIT: The impression I get from around the internet is that because so much is shared between the two cores in each module, this should really be considered a quad core in many tests. On the odd occasion that all 8 cores can be made use of properly, this thrashes everything out there (apart from the 6C/12T ofc) - so I'm guessing that this might be something of a multitasking beast. It's also highly inconsistent even when these core use issues don't come into play. Sometimes it's right up there (although HH's gaming results are a tad unusual and perhaps limited to a few games) but at other times it's not even close - occasionally even seeming to fare worse clock for clock than 65nm Intel quads released nearly 5 years ago.
     
    Last edited: Oct 12, 2011
  5. Stuart_Davidson

    Stuart_Davidson Well-Known Member

    Joined:
    Nov 20, 2002
    Messages:
    5,843
    Likes Received:
    188
    Trophy Points:
    73
    Thats pretty much what we were getting at in the article, this is a CPU which is designed for future perfomance than todays. To be honest AMD would have been happy to have everyone review with a selection of hand picked benchmarks which take advantage of the multi-threaded perfomance (and new instructions) of the FX and do it on Win 8... but that doesnt give a fair representation of what the average user will get.

    For the games, i'm confident that our results will be more correct than other sites. Various reasons for that, many of which involve the tests/practices other sites use, so I wont go over that all again. Overall though any recent game which is impacted by CPU perfomance and cores will do well on AMD. F1/Dirt3 are examples of this and pretty much any decent engine going forward. (Custom engines also usually support more cores than older licenced ones, a good example being Star Trek Online which uses at least 4 cores... try getting a repeatable bench of a MMORPG though, not going to happen!)
     
  6. blibbax

    blibbax nahm8

    Joined:
    Sep 6, 2008
    Messages:
    6,210
    Likes Received:
    203
    Trophy Points:
    73
    The performance victories in games were pretty slim, though. I really wanted to see this CPU do well, but right now it almost looks as if AMD's best move would be to start shipping higher-clocked die-shrunk Phenom X6s (and maybe 8-12 core Opterons for AM3+) and abandoning this new architecture altogether - to go from 45nm to 32nm and lose both single threaded performance and power efficiency is ridiculous.
     
  7. eblg73

    eblg73 New Member

    Joined:
    Oct 12, 2011
    Messages:
    10
    Likes Received:
    0
    Trophy Points:
    0
    Seems you guys are the only website on this whole planet that recommend this new CPU. I wonder why...

    It's hardly faster than the previous generation 1090/1100T in most applications. Game tests are run here at AMD specs of the reviewers guide, so plz don't start with the AMD favoured benchmarks, coz that is exactly what you guys did !! Lower the res and exclude the GPU and see what gives...

    To sum up Bulldozer : IPC is bonkers, powerdraw is insane... it will work well with some multithreaded applications ( with special instructions for the BD ) and to top it all off release prices are too high to even compete with a 2500K setup...

    I wonder why there are only recommend and top notch hardware here...
     
  8. Alereon

    Alereon New Member

    Joined:
    Oct 12, 2011
    Messages:
    2
    Likes Received:
    0
    Trophy Points:
    0
    I want to express how concerned I am that Hardware Heaven only published benchmarks where Bulldozer at least did adequately, and none showing the poor performance users can actually expect. As a result the product was awarded a 9/10 rating that is certainly not deserved. At best, this shows that poor attention was paid to benchmark selection in order to get a good cross section of performance. Less charitably, this shows a disturbing lack of editorial integrity.

    Readers rely on reviews to give an accurate, unbiased picture of a product's performance and how it stands up to the competition. If you look at essentially every other review on the web, you see that performance is characterized as disappointing to abysmal, and Bulldozer is not recommended. See the conclusion to the Anandtech Bulldozer review for example. I'll also point out that the reviewer called out Bulldozer for needing a lower price, yet still awarded a 9/10 for value. While you're certainly allowed to have your own opinions, giving a product like this a 9/10 rating is a huge disservice to your readers, and forces the conclusion that your reviews are written for the benefit of your sponsors. That's a harsh charge I know, but one that I feel is warranted after reading the tone of the conclusion, the way the products flaws were glossed over (no mention AT ALL of the high power usage), and the way the numerical ratings fail to reflect either the content of the review or reality.
     
  9. IvanV

    IvanV HH Assassin Guild Member

    Joined:
    Dec 18, 2004
    Messages:
    10,792
    Likes Received:
    2,004
    Trophy Points:
    138
    Well, they couldn't have been much more than slim, considering that on those settings the games were more GPU than CPU limited. A more artificial situation (low resolution) where CPU would be a bottleneck would better show off the difference in speed with which the two CPUs process game data. I would like to have seen such tests alongside the existing ones (which represent current realistic real world scenarios), for "academic" purposes and as an indication of how the CPUs will stand in the future when engines and games become even more CPU intensive.

    BTW I feel like the review could have been a little more in-depth. The CPU was properly run though its paces, but looking at the whole text, despite this being the launch review for a whole new CPU architecture, the processor wasn't given any more space than what those which represent mere speed bumps of well known predecessors receive.
     
  10. Liqourice

    Liqourice Well-Known Member

    Joined:
    Nov 16, 2005
    Messages:
    3,837
    Likes Received:
    339
    Trophy Points:
    93
    Isn't it fascinating that there's always 1-posters that come in here with such hard critics?

    I'm truly disappointed with how this CPU performs, and I don't think for one second that the review here is biased in any way since it clearly shows that it doesn't perform as well as expected. To you two critics ( I wouldn't be surprised if you're one and the same) I can only say: What makes you so sure that other reviewing sites aren't biased in the other direction?

    I only need to read this review to know I will most likely not buy one, even though I'm an AMD fan. I got the information I needed to make that decision from a review you claim to be biased.
     
    YnHDaRkMaN likes this.
  11. Alereon

    Alereon New Member

    Joined:
    Oct 12, 2011
    Messages:
    2
    Likes Received:
    0
    Trophy Points:
    0
    I'm a 1-poster because I was disturbed enough at this review to register and post about it, which I think speaks to the gravity of the situation. I am not "eblg73", as the difference in tone and writing styles should make clear.

    You say that you're not going to buy a Bulldozer, despite being an AMD fan. Can you honestly tell me that you believe this product deserved a 9/10 rating? How would they have rated Bulldozer if it soundly beaten the competition, 15/10? I don't think I need to touch on how ridiculous the assertion that "maybe everyone else is just biased the other way" is, but I'll once again remind you that those sites based their conclusions on a wide variety of tests, while this site only published the tests where Bulldozer doesn't look bad.
     
  12. skylineaddict

    skylineaddict Sleep Deprived Ant

    Joined:
    Aug 13, 2008
    Messages:
    944
    Likes Received:
    102
    Trophy Points:
    0
    I, unfortunately, have to agree with Alereon. He is a little harsh with the criticism, but he's right when he says that the rating at the end of the review doesn't quite match the observations made during the review.

    As for Bulldozer itself, I am thoroughly disappointed in AMD. But, redemption may or may not lie with Windows 8 and/or the new instruction sets.
     
  13. blibbax

    blibbax nahm8

    Joined:
    Sep 6, 2008
    Messages:
    6,210
    Likes Received:
    203
    Trophy Points:
    73
    I can kind of understand the 9/10 rating. Yes, this product doesn't quite square up to the competition, but it's still a good product. It's not actually lacking, it's just not as good as its (fantastic) peers. If the 2500k and Phenom II X6 were 30 or 40% dearer, this would be a good buy.

    That said, if they'd released this in 2008 I feel we'd still have been a little underwhelmed, so maybe I'm wrong.
     
  14. Liqourice

    Liqourice Well-Known Member

    Joined:
    Nov 16, 2005
    Messages:
    3,837
    Likes Received:
    339
    Trophy Points:
    93
    I think people put to much value in the final rating. As blibbax says it's not a bad product and for most people it'll be a perfect buy. For those of us that care enough and want the most out of our money it doesn't really matter since neither of us really cares about the rating, we look at the test results and value that in comparison to what we expect.

    Reviews aren't only made for the high-end users, they're made for all kinds of people and for the average user a 9/10 rating is perfectly fine.

    I can't agree that the posted results are only ones where the AMD doesn't look bad. I think it looks really bad in some of them. It all depens on how you chose to read the results.
     
  15. Judas

    Judas Obvious Closet Brony Pony

    Joined:
    May 13, 2002
    Messages:
    38,931
    Likes Received:
    1,116
    Trophy Points:
    138
    while i'm happy.. i'm not entirely impressed..

    Considering what amd is going to be heading into in the next few months.... i expected a significant change in their design in terms of their memory controller.

    There is clear advantages for example in triple channel memory solution.... and now that we see roadmaps with quad channel for intel.. this will really leave amd in the dust on the memory front.

    AMD used to be a leader in the memory.. being the first to make heavy use of the onboard memory controller and dual channel.

    Now they are trailing.

    I'm happy to see that "some" of the benchmarks show the amd cpu matching up or showing a margins of error lead. But it's just not significant enough.

    I also know that above 4 cores many programs aren't really showing any advantages in the benchmark area.

    I don't know.. but i was really hoping that AMD would have something that could contend or at least try to contend with the Socket 2011 parts that are expected soon.
     
  16. eblg73

    eblg73 New Member

    Joined:
    Oct 12, 2011
    Messages:
    10
    Likes Received:
    0
    Trophy Points:
    0
    2 posts now.

    Explain value 9/10 : 2500K pricing is far lower then the FX-8150. Yet the FX-8150 has a really hard time with the Intel SB competitor.On top of that to get good performance you need a decent 990FX motherboard. Not cheap either... only way to let the value claim look good is to compare it with Gulftown. This CPU doesn't come even close to that hardware. I love to see how many boards PWms will pop when running these CPUs at 5Ghz daily...nice recommendation btw...

    The reviewer has followed 100% AMD's guidelines in games to explicitly show good performance. Advised : Cranck up the detail and res till the CPU is ruled out and the GPU is practically on it's knees. Excellent methodology to test a CPU. Not...

    Tip : Next time select Unigene, it almost rules out CPU completely, that might be a good CPU test then.


    If any of you want a copy of how you should run the tests to make the CPU shine I can send it to you, you would be shocked at the similarity in test selection and output. I spend 5 days and nites with BD and had to conclude in the end, sorry AMD try again in 2012.

    Did I see a post about other sites being Biased ? Plz have a look at this site's rating/awards : not much sub 8/10 out there. Unless something has recently changed ! Doubt it though

    Quick overview :

    Processors:

    - AMD Phenom II X6 1100T Black Edition (9/10, silver award)
    - Intel Core i7-950 (10/10, gold award)
    - Intel Core i7-2600K Processor (10/10, gold and performance award)
    - Intel Core i5-2500K Processor (10/10, gold award)
    - AMD Phenom II X4 975 Black Edition (9/10, silver award)
    - Phenom 2 X4 980 Black Edition (9/10, recommended)
    - AMD A6-3650 Accelerated Processor (8/10, value award)

    Mainboards:

    - Intel DP67BG Motherboard (9/10, performance award)
    - Intel DH67BL Motherboard (9/10, silver award)
    - ECS A890GXM-A2 Black Extreme (9/10, value award)
    - Sapphire Pure Platinum A75 (9/10, recommended award)
    - Sapphire Pure Platinum H67 (9/10, recommended award)
    - ASRock Fatal1ty P67 Performance (9/10, performance award)
    - ASRock Z68 Pro3 (9/10, value award)
    - 990FXA-UD7 (10/10, gold award)
    - Gigabyte A75-UD4H (9/10, recommended award)
    - Maximus IV Gene-Z (10/10, gold award)
    - Fatal1ty Z68 Professional Gen3 (10/10, gold and performance award)

    Power Supplies:

    - Silencer 760W (10/10, gold award)
    - OCZ ZX Series 850W (9/10, silver award)
    - Corsair GS800 (9/10, silver award)
    - High Current Gamer 900W (9/10, value award)
    - Corsair HX1050 (9/10, performance award)

    Graphics cards:

    - Sapphire 6950 DiRT 3 Edition (10/10, gold and value award)
    - ASUS Matrix GTX 580 (10/10, diamond award)
    - Powercolor Radeon 6950 PCS+ Vortex 2 (10/10, gold award)
    - HIS Radeon 6970 IceQ Mix (9/10, recommended award)
    - NVIDIA's GTX 580M (9/10, performance award)

    Memory:

    - Kingston HyperX DDR3-2400 (10/10, gold and performance award)
    - Kingston HyperX DDR3-1600 LoVo (10/10, gold award)
    - Corsair Vengeance 12GB (10/10, gold and value award)
    - Corsair Vengeance DDR3 Kit (9/10, value award)
    - Crucial Ballistix Smart Tracer DDR3 Kit (9/10, recommended award)
    - G.Skill Ripjaws DDR3 Kit (9/10, recommended award)
    - Kingston HyperX t1 DDR3 Kit (9/10, recommended award)
    - Patriot Viper Xtreme DDR3 Kit (9/10, performance award)

    Spot a simularity ?

    Conclusion of the day :
    If you want to make sure you are getting a good product : it has to be tested by Hardwareheaven.net. Coz everything is well under the sun there. Awards are handed out by the masses and it seems the readers dig it, coz it all looks kindda fancy...
     
    Last edited: Oct 13, 2011
  17. mkk

    mkk Well-Known Member

    Joined:
    Oct 9, 2003
    Messages:
    5,334
    Likes Received:
    152
    Trophy Points:
    73
    I don't care what people think about any sites reviews, but I've noticed today reading reactions all around that a lot of people aren't really reading what even the review summary says but goes mostly by awards or ratings. That's a bit of a problem with having numericals or awards, but it's mostly the fault of the reader.

    About these CPUs I'm not really disappointed with the performance given what we've already known about the architecture for a good while. What made me cancel my pre-order today was the power consumption figures. Had the FX-8100 model been out on release at a nice price I might still have given it a shot, but the 8120/8150's aren't what I had hoped to be powering my system with for the next several years.
     
  18. eblg73

    eblg73 New Member

    Joined:
    Oct 12, 2011
    Messages:
    10
    Likes Received:
    0
    Trophy Points:
    0
    That's just my point, the rating/awards are not reflecting the articles contents... is that the readers fault or the reviewer biasing ? I see it more an act to satisfy the PR and marketing people. They will interpret the rating correctly as they usually don't read the article and go straight to the bottom of the last page !! Why ? Coz they love awards !!

    How can you give 9 value when you tell the CPU's price is too high a few lines above.. Don't grasp that at all.... The CPU is just too expensive for the way it performs. It should be priced around 2500K level... then it will sell...

    But it seems comon stuff here that the ratings are well overrated...

    ...A mobo that scores 10/10 is the perfect motherboard, sorry but that does NOT exist.... it has flaws, overpriced, clocks like crap,... 10/10 I want to test that one...

    So you knew the architecure had shortcommings, so that makes it okay the way it (under)performs in many applications ? It puzzles me the logic behind some reasoning here...
     
    Last edited: Oct 12, 2011
  19. jandarsun8

    jandarsun8 Active Member

    Joined:
    Jan 2, 2008
    Messages:
    459
    Likes Received:
    30
    Trophy Points:
    38
    Good grief, who the hell cares about what award was actually given out at the end of it. People are smart enough to look at the over all article and make a decision for them selves.

    What I do see here is two people that came on here to criticize the test being ran and what award it got and that it looks like two Intel employees looking for raises. If you didn't agree with how the test were set up, that's fine and one thing. Post how you disagree with it but to go into such detail as to why, what was wrong with it, and that the whole site sucks, then just leave.
     
  20. eblg73

    eblg73 New Member

    Joined:
    Oct 12, 2011
    Messages:
    10
    Likes Received:
    0
    Trophy Points:
    0
    One) I want to understand the true logic behind the rating system. As the rest of the contents is usually okayish. But why claim A and rate B. You don't care about rating, well I do sir. So do they in Taipei and Canada.

    Two ) And a reply from the reviewer why he stated that he did it better than all the rest as his game scores seem to be more correct. You have got no idea how many threads/discussions there are concerning this particular claim.

    I'm not payed by any company and wished this CPU was a big hit. Competition is beneficial for the customer...
     
Thread Status:
Not open for further replies.

Share This Page

visited