http://arstechnica.com/business/news...t-kill-gpu.ars
Interesting take on the future of CPU/GPU dynamics and which will be more important in the future.
7 years ago I sat in a Nvidia company presentation where the Nvida rep (one of the R&D leads) made a comment that with how large GPUs are, it would only be a matter of time before CPU+GPU ship was released.
Kinda funny, IMHO, if AMD/ATI beats them to it.
Well, AMD beat them to it, not ATi. That's a big difference, especially considering that nVidia has no x86 license.
The thing is, AMD is at least a generation behind nVidia when it comes to GPU compute and have been for sometime. Not just hardware, but the complete ecosystem.
Oh, and not to mention that Fusion chips still aren't out and the 'real' Fusion (with a serious, full power x86 unit) has been delayed into the middle of 2011.
We've been talking about this for years. Reality is that it's cheaper and ends up making better business sense to keep them separate. Nvidia's IGP's have gotten really good. You can play DX 9 games with high settings on dirt cheap IGPs. Vendors can mix different IGP/dedicated solutions across their models easily, and cheaply.
In order for something like this to ever really take off it's going to have to be a lot better and a lot cheaper. You'll need a large player like Dell, HP or Apple to back this across a number of different models before they can reach any kind of scale that makes sense.
I think the best place for a product like Fusion are applications extremely sensitive to heat and power. Obviously that means laptops, but I think an architecture like this would be quite useful for HPC applications. There are, however, a couple of problems for AMD when it comes to using it's Llano-type processors in supercomputers.
1. They aren't even building versions with multiple HT links to my knowledge. I've never really heard of a serious supercomputer being built out of single socket nodes. When you're trying to cram in as many cores as space and power will allow, using multiprocessor nodes has an obvious advantage.
2. nVidia eats their lunch when it comes to using GPUs as HPC accelerators. Not only are nVidia's GPUs better performing from a DPFLOPS/watt basis, nVidia also actually has a software framework. AMD continually talks about this or that (Brook+, CTM, now OpenCL) but never really gets anything done.
Those two obstacles are going to stop someone like Cray, SGI, or IBM from using thousands upon thousands of these chips in HPC systems. Well, that and the fact that SGI has pretty much gone with intel, IBM is now split between POWER7 and intel, and it's very likely Cray will also go with intel once it's series of supercomputers developed from Red Storm (XT3, XT4, XT5, XT6) reach the end of their life.
Surely you jest?
9400M gets smoked by $40-50ish 9500GT and HD4650. You can now grab a 9800GT, HD4850, HD5750, or (on recent Newegg coupon deals) an HD4870 or 9800GTX+ for $100 - all of which are in a completely different league than the 9400M.
Oh yeah, and the best CPU you can run it with is a C2Q.
That leaves us with the HD4290 as our high end IGP. It generally costs $30-40 more than mobos with the somewhat slower HD4200 (although you do get some other benefits from the upgrade) and even the 4290 is slower than 9400M.
It's better than 9400M in that you can use it with Phenom 2 instead of C2Q, but it's slower than a current gen entry level discrete card.
On what planet is either 9400M or HD4290 even close to a replacement for any of the cards I listed?
Unless of course you're referring to rumors that mainstream Fusion will be carrying HD5650-equivalent graphics on-die... But of course that's not a mobo GPU, and even then a 5650 doesn't replace a 4850 or a 5750 or a 9800GT.
Intel and AMD are both betting against you (note that Sandy brings graphics onto the CPU die as well).
Unless he means intel's integrated "GPUs". Those are the most common graphics adapters out there.
If he means stuff like nVidia's integrated solutions or AMD's, I'd like to see evidence.
How are Intel GPUs possibly a replacement for an HD 5450, much less an HD 4850? That makes even less sense.
Ima stated that mainboard IGPs are replacing $75-100 PCIe cards. I can't think of any context in which that hold true. The pinnacles of mobo IGP are the aforementioned HD 4290 and 9400M, which are the equivalent last-gen entry-level $30ish PCIe cards at best.
Even Sandy Bridge's impressive graphics (per Anandtech's preview) are only pushing into the $50 space. And, of course, the clear future of IGPs is in CPUs a la Sandy and Fusion. Mobo IGPs will be a distant memory a year from now.
Yeah, I really don't know. I don't pay attention at all to OEM desktop computers, I pay a glancing bit to workstations but they always have some form of discrete card.
Then you have laptops, which are an entirely different ballgame. The differences between on-board and discrete are pretty slim there.
Notebooks (and netbooks). You know, the largest, fastest growing segment of mainstream computing.
Most gamers don't play at 1920x1200 with all settings set to ULTRA and 8xAA/16xAF. Less than 20% play at 1080p or higher.
IGPs can now run most games at 720p resolutions with medium settings and 2xAA/4xAF. 60% of gamers play at resolutions 1280x1024 and below. NVIDIA's IGPs are absolutely KILLING it in the mobile market right now. A few years ago playing a game on an IGP was kind of a joke. Now it's completely legitimate.
You can play Portal on 13"-15" notebook, at its native resolution, with medium/high settings and basic AA/AF, at 45+FPS. That's remarkable, and a HUGE leap from where we were just a few years ago with garbage like the Intel GMA 950. That's probably around the time that AMD thought this Fusion gimmick was going to be a big win—back around the time the Intel 915G/945G chipset was the most popular thing since... maybe since ever. Seriously, the 945G may be one of Intel's time best-selling platform chipsets. That thing is everywhere.
The 9400M has changed the entire landscape for IGPs and mobile graphics.
I'm a bit biased though, since I'm not even really interested in the big-gun cards that play games at 4K resolutions with full AA/AF. I don't have a 4K screen, and neither do most people. $500 for a video card? Meh. I'm interested in when we'll have $500 boxes with IGPs that can play Portal 2 at 1080p. And really... we are like a year, maybe two, from that. NVIDIA and Intel will own that space. No one is going to want this AMD Fusion garbage.
Seems like a pretty early dismissal given that
1. AMD's approach is highly similar to Intel's with Sandy (is nobody going to want Sandy either?) Intel just beat them to it with a heat start on process tech and engineer power. The fact that Intel is following AMD's footsteps (in realizing where tech is going) should be good evidence that Fusion is actually a good idea.
2. There are massive technical benefits to having the IGP on-die rather than on the chipset. Far lower latency to both CPU and memory, L3 cache access as per Sandy and probably Fusion, no need for an ultra-wide system bus to manage the IGP, being manufactured on a state-of-the-art process rather than on one 2-3 generations old as chipsets tend to be, saving power via advanced gating and other technologies, more efficient thermals via advanced manufacturing process and having a single hot area that needs cooling, etc.
The efficient thermals are proven with Sandy's combination of an IGP that absolutely destroys the 9400M and a 95W TDP (remembering that Intel's TDP estimates are a theoretical worst-case scenario and realistic usage even under max load tends to be substantially lower).
Given all that, and given your own arguments about the importance of IGP performance, I don't see how you can possibly trash on Fusion. It's an amazing way to increase IGP performance, it'll reduce chipset and bus complexity and lead to smaller and more efficient form factors, and it's one more step toward mainstream SoCs and ITX-sized computers with ATX-sized performance.
Even if you're skeptical of AMD for some reason, are you really going to say nobody will be interested in Sandy either? It's better than Core2+9400M in every way you can name, and even if nVidia somehow gets a bus license from Intel, their chipset IGPs will be at a huge technical disadvantage being made on older manufacturing processes and not having L3 cache access or ultra-low latency to CPU and memory.
I completely agree though about $500 video cards being ridiculous. I'll never own another desktop or another laptop with discrete graphics (my last laptop had a 2600XT). Heat, battery life, and price all heavily favor IGPs, and with both Intel and AMD putting IGPs on-die you can bet that's the way the market is going to go.
So why pick on Fusion when Sandy is so similar?
You might be disappointed with Llano being a K10 derivative (no, it's not a K8 regardless of what Fuad quotes in that conference call... AMD was referring to the fact that K10 is a K8 derivative, not that they're actually going to use the K8 core - in all likelihood it'll be a process shrunk quad core version of Thuban with the usual tweaks) but you already said you're not a desktop guy. Ontario might be too low end for either of us (although there's a chance I'll get it if Sandy is too expensive and/or it rivals Sandy's IGP and the P320's CPU) but the concept is gold and I think Bulldozer+Fusion will probably be a solid price/performance candidate in the laptop space.
Intel following in AMD's footsteps? Intel has been leading the IGP push for the last decade. Nvidia got in the game and forced them to push forward a little harder. Sandy is the evolution of Intel's own IGP platform. AMD went like 6 years without selling a platform. It wasn't until they bought ATI that they even tried. They bought ATI who has managed to work together with AMD to make some very mediocre, short-run products. Seriously, is there a single OEM using the AMD platform?
And that's why Fusion doesn't even make any sense. Who [OEMs] is going to buy it? On what platform? What kind of products can OEMs even create? If AMD was Intel and their platform was the standard, they could pull it off. They could create dozens of different SKUs and drop them in. Intel can get a single platform into market and keep it there for four years at at time.
Fusion is AMD following in Intel's footsteps, ignoring the fact that they don't have a platform or the clout to pull this off. They can't get the pricing. They won't even be able to offer this at a highly profitable margin, because they can't do the volume.
Here is my review without even using it:
Balls: 10/10
Effort: 8/10
Execution: 2/10
Timing: 1/10
AMD is killing itself. AMD is still building products while Intel is building solutions.
Aside from Gateway, Lenovo, Toshiba, Acer, and HP you mean? Well, there are probably others but I'm too lazy to dig up a list. Check out AMD's platform partners. By my count they're missing Dell (always the Intel diehards) and some of the small companies.Quote:
They bought ATI who has managed to work together with AMD to make some very mediocre, short-run products. Seriously, is there a single OEM using the AMD platform?
And AMD's platforms are mediocre and short-run? HD3200 had a solid year of well-reviewed life in the budget sector, HD4200 has built on the success of HD3200 and combined it with decent CPUs (which is really what AMD lacked in the mobile sector since nVidia IGPs were quite a bit more expensive than AMD and not a whole lot better until the 9400M) like the M300 (only drawback is 35W ACP but mine has never had heat issues even during long SC2 sessions). Now the HD4200 refresh (HD4250/75) is doing the same with the 25W P320 and related processors and finally has most of the major OEMs signed up and producing kit.
Ontario will probably be a hit for outperforming Atom at similar price points (and probably performing much like a P320/HD4250 in a far lower ACP/TDP). Llano will be a tougher sell but it will also have a pretty short market life; looks like 6-9 months.
You still haven't addressed my points about the technical superiority of on-die graphics. They can certainly be disabled if nobody wants them but like you said, discrete mobile graphics tend to be an overpriced, overheated sadgasm that kills mobility. On-die increases performance and decreases heat and power. How can you possibly be against that?
Why is it a gimmick when AMD does it and just fine when Intel does it? Because AMD doesn't have Intel's brand presence? That makes no sense. How is selling a CPU with on-die graphics any harder than selling a CPU with mainboard graphics? Especially when the on-die graphics get a free performance upgrade at better power and heat levels.
The current AMD platform: criticize it if you will, but it definitely exists and it definitely has the majority of major OEMs signed up for it. Note that the system in my sig is part of the Dragon *platform* as well. http://www.amd.com/us/products/noteb...-platform.aspx
Oh yeah, they're missing Asus too as far as I know. But Sony has an extensive AMD line so I'd say that's a tie. Sony is overpriced, but they aren't exactly known for selling laptops based on mediocre, short-lived platforms. Remember their silence on AMD during the crappy Turion days? Something convinced them to warm up to AMD since then, and with a pretty wide range of models using pretty much all of AMD's mainstream platform options.
And there's rampant speculation that even Apple will sign up for Ontario in the next iPad and possibly Mac Mini/Macbook. iMac is basically a desktop with no high-end graphics option so that'll probably carry Sandy instead.
AMD sells a seemingly random assortment of chips to most OEMs. A mix of what seem to be constantly rotating chipsets and CPUs, whatever is cheapest at the moment.
Intel ends up singing long term deals with OEMs for chipsets, and in some cases actual motherboards (often through partners like Foxconn). It's one of the reasons that AMD is pretty much absent from IT. I can still easily get a new Intel 945G motherboard, even though it's 4-years old. Hell, I can probably buy it boxed, retail, directly from Intel.
My point is that in order for something like on-die graphics to work, you have to sell the complete solution; and you have to sell it at a scale that makes it profitable. If, all of a sudden, every notebook has a Sandy-equipped motherboard then it's easy for OEMs to order a wide range of chips and drop them into different SKUs. Thus, due to the volume, they are cheap enough to be economical for the consumer.
If, on the other hand, every time an OEM wants to create an AMD Fusion product they have to order all new components, often from a different manufacturer, the result is a non-price competitive product.
Even if Fusion is a good performer it won't be able to be truly price competitive in the way AMD products need to be (i.e. priced lower than an Intel product). And OEMs won't adopt it because they don't have a sustainable platform to build from.
Sandy is the evolution of Intel's IGP. For the last 4 years they've been selling complete solutions. When OEMs wanted to create something with a little more punch, they threw a dirt cheap NVIDIA IGP against it. Now Intel is moving to a new platform.
AMD, on the other hand, has spent the last 4-years struggling to incorporate ATI into their portfolio. They have no platform, and while ATI was once the IGP leader, they have now fallen into 3rd place.
Sure, AMD has pretty fast turnover, but they don't sell random chips. They sell the newest generation and the generation previous to it. That may be a problem for IT, but AMD parts are generally quite cheap and their hardware has a long lifespan. Sure, you may be able to buy a new 945G direct from Intel, but can you plug your Core2 processor into a Lynnfield mobo? How about plugging your Nehalem into a P45/55? And Sandy will require *2* new sockets.
On the AMD side, I can take the old X2 6000 someone gave me as thanks for helping them reformat their hard drive and plug it into a shiny new 880G mainboard for an ultra-cheap desktop with very decent multimedia capability. I can buy a new X4 945 and plug it into some old AM2+ mainboard, holy cheap upgrade Batman! It just works! I don't need to buy new parts from Intel when I can mix and match old and new parts as I wish. That's a much better solution.
Also, AMD can easily make up for their poorer economy of scale with more down to earth pricing. Even now they rule the sub-$100 market and have little performance competition below the $200 i5 760. Phenom X4 kills Clarkdale in everything except single-threaded performance; Clarkdale is a decent gaming/HTPC solution but that's about it. Problem being not that Clarkdale is bad but that it's priced a quarter higher than it should be.
AMD is fine as long as Intel's looking for inflated margins, and Bulldozer should be enough to make them competitive again, if only up through the midrange. And the on-die graphics will give AMD a much-needed performance boost. Without it they'd basically become the next Intel without embarrassingly poor graphics that hold back the industry... but none of Intel's market power to force their junk down everyone's throats.
Intel will already kill off AMD's entry level discrete cards and utterly destroy its mobo IGPs with Sandy. Fact is, regardless of IT acceptance (where margins are tighter and Intel's got a massive upper hand anyway), AMD *needs* Fusion to compete.