Discussion Forums > Technology
Your view on AMD's Bulldozer
ColdFission:
Quickly read the conclusion from bit-tech (great UK publication too) and they don't hold back.
--- Quote ---The final grievance is the price of the FX-8150 – at £205 it’s £37 more expensive than a Core i5-2500K and only £40 cheaper than a Core i7-2600K. It fails to outperform either conclusively, with only some victories over the quad-core i5-2500K in some heavily multi-threaded tasks. But if you’re after multi-threaded performance, the £40 extra for a Core i7-2600K is more than justified. Perhaps the 6- and 4-core flavours of the FX will prove interesting upgrades, but the FX-8150 definitely isn’t.
We therefore feel totally vindicated that at no point did we recommend any bit-tech reader to buy a Socket AM3+ motherboard ‘to get ready for Bulldozer.’ We merely reviewed these boards on the premise that they were new and that people might wish to buy one as an upgrade for a Phenom II system – we had no idea whether Bulldozer would be good, bad or indifferent, so we urged caution. Turns out we were right: the FX-8150 is a stinker.
--- End quote ---
The overall feeling from various sites is that, it just sucks. This isn't the Intel killer like the FX-57 was all those years ago that many in the AMD fanbase were hoping for. Personally, AMD needed a win not just for it appearance in the desktop market but for financial reasons. I stresses in my OP post that AMD can't screw up like they did with Barcelona. Well, looks like they pulled another "Barcelona" and crapped all over themselves. However, I suspect that Bulldozer variants known as Interlagos and Valencia, the server parts, will do very well in those markets. Running hundreds or even thousands of VMs, running apps that spit out hundreds or thousands of threads 24/7, HPC apps, etc should make the server parts shine.
I guess what people forget is that Bulldozer was and is a server orientated part first, desktop part a very far second. And no one can really fault AMD for that thinking because the server market is a place with high margins and this where AMD needs to make some wins here. As for the desktop front, I never thought the Bulldozer was going to be the Intel killer. In my OP, I did say:
--- Quote ---Well, as for my personal view on how it will perform for the desktop side of things, I think it will just do fine competing with Intel's current offerings. Is it going to be like the old FX in the Athlon 64 days? No, I don't really think so. I mean, this was supposed to be released back in late summer but was delayed and delayed. Previous rumours suggested a September 19 launch but that has come and gone. There have been issues with the brand new 32nm process from GlobalFoundries (where the Bulldozer chips are made, AMD is a fabless company in which they just send their designs to them, a point I forgot to mention earlier on) so it may impact performance expectations.
--- End quote ---
But, there is still light at the tunnel and has been mentioned already and its Trinity. Again, Trinity will be their new Fusion APU combining AMD's Piledriver architecture (Bulldozer "enhanced") and their new graphics architecture known as GCN (Graphics Core Next) which will also be used in their upcoming HD 7000 series. In my view, this will do very well in the mobile as well as in the mainstream desktop markets as this will provide more than enough CPU grunt for most people and as well as more than enough GPU power to playback HD films and play some games.
Overall, I think AMD is going to do the same process like they did with the first Phenoms. As time went on, they released revised steppings that lead to the Phenom IIs which performed quite well for their targeted markets and were much faster than the first Phenoms. So, expect newer steppings to come early next year, they will be better but I don't think it can wipe the tarnish of this very poor launch.
kureshii:
ColdFission: Not really. Don't expect Trinity to turn things around for AMD either. Although it will hopefully keep them in business for the time being.
For one, Trinity will only have 4 Piledriver cores (2 modules), which they claim will get a 10% per-core performance improvement over the Zambezi cores. That's still pretty disappointing performance.
For another, GCN isn't going to appear in 7-series Radeons. It's not even mentioned on their roadmap, which stretches to 2012, which means we'll see it in 2013 at the earliest.
Only thing to look forward to now is Opteron server benchmarks and see if it justifies the poor desktop performance.
ColdFission:
--- Quote from: kureshii on October 12, 2011, 04:41:28 AM ---ColdFission: Not really. Don't expect Trinity to turn things around for AMD either. Although it will hopefully keep them in business for the time being.
For one, Trinity will only have 4 Piledriver cores (2 modules), which they claim will get a 10% per-core performance improvement over the Zambezi cores. That's still pretty disappointing performance.
For another, GCN isn't going to appear in 7-series Radeons. It's not even mentioned on their roadmap, which stretches to 2012, which means we'll see it in 2013 at the earliest.
Only thing to look forward to now is Opteron server benchmarks and see if it justifies the poor desktop performance.
--- End quote ---
Well, I do see Trinity as more of a success than failure as it will provide more than enough CPU resources for a majority of people for their daily tasks. Sure, it won't all the benchmarks, but fused with a great GPU architecture, will spare the regular user from investing in purchasing a graphics card. Trinity won't be targeting the enthusiast or the gamer, but the majority (where AMD is should make the most money from) who are people who like to play games on occasion, e-mail, word processing, like to watch a lot of HD content, web browsing, etc. From what I read from the mobile Llano reviews, many of them were positive pointing the great graphics performance for a mobile platform compared to what Intel has for their GPU. Llano is slower on the CPU side, then again, more than fast enough for a majority of users out there. As for the 10% performance increase, IIRC, that was a slide from here:
FX Next
As for GCN, there were some articles that I have come across saying that the HD 7000s will use GCN (I know there are no roadmaps out there right now detailing that).
http://www.xbitlabs.com/news/graphics/display/20111007144928_AMD_Demos_28nm_Graphics_Solutions_Again.html
--- Quote ---The 28nm generation of AMD's graphics processors will be rather broad. In fact, it is rumoured that even within Southern Islands family there will be chips with VLIW4 architecture as well as more progressive so-called GCN (graphics core next) architecture.
--- End quote ---
http://semiaccurate.com/2011/08/30/trinity-rumors-confirmed-gpu-speeds-and-more/
--- Quote ---hat leaves the GPU. If you notice, the GPU is listed as HD7000, aka Graphics Core Next (GCN), aka Southern Islands. That means going from VLIW5 to scalar + VLIW4, whatever the code word for that is. In any case, going from 80 ‘old’ clusters (400 shaders) to 120 ‘new’ (480) clusters is where the majority of the 50% comes from. Throw in an updated memory controller, tighter integration between the sides, and you have not only more speed, but much more exploitable speed.
--- End quote ---
Well, I (and maybe these articles) could be wrong anyway as no of this information is official. Anandtech remarks that GCN won't be used with the next-generation of APUs:
Anandtech
--- Quote ---Because of this need to inform developers of the hardware well in advance, while we’ve had a chance to see the fundamentals of GCN products using it are still some time off. At no point has AMD specified when a GPU will appear using GCN will appear, so it’s very much a guessing game. What we know for a fact is that Trinity – the 2012 Bulldozer APU – will not use GCN, it will be based on Cayman’s VLIW4 architecture. Because Trinity will be VLIW4, it’s likely-to-certain that AMD will have midrange and low-end video cards using VLIW4 because of the importance they place on being able to Crossfire with the APU. Does this mean AMD will do another split launch, with high-end parts using one architecture while everything else is a generation behind? It’s possible, but we wouldn’t make at bets at this point in time. Certainly it looks like it will be 2013 before GCN has a chance to become a top-to-bottom architecture, so the question is what the top discrete GPU will be for AMD by the start of 2012.
--- End quote ---
With, this quote, you would be correct. But again, none of the information about Trinity or the upcoming HD 7000s series is official. However, AMD did do a Fusion Developer's Summit trying to push their APUs to devs and showing how much better their programs and what not would run on AMD APUs (its a really crappy description, but that's the gist of it) because of the highly parallel side of the GPU. IIRC, it also introduced GCN in that summit. To me at least, it would make sense to pair GCN with Trinity. But again, its just speculation on my part.
Anyway, I think I'll be going Intel (again) for my next PC overhaul that I hope to do early next year.
nstgc:
Bulldozer failed to hit the mark (I'd call it "crash and burn personaly")
http://www.xbitlabs.com/articles/cpu/display/amd-fx-8150.html
vuzedome:
After going through several reviews, personal frequents, all I can say is I am disappointed.
I thought I can get something nice maybe by Christmas but sadly, will have to wait for IVY.
Navigation
[0] Message Index
[#] Next page
[*] Previous page
Go to full version