ATI puts Stream everywhere
IF YOU THOUGHT today was only going to be about AMD CPUs, the company has a streaming surprise for you. Up until now, it was pretty quiet about the strategy, but by this time next month, Stream will be everywhere.
The big bang is that Stream consists of three major parts, the GPU, the Compute Abstraction Layer (CAL), and tools. You know about the GPU, and tools are just that, tools for writing code, not really end user focused. The big one is the CAL.
Normally, with something as esoteric as GPGPU, you need to download drivers, patches and software, followed by an afternoon of tuning. It puts the technology in the proverbial geek ghetto, not something for the faint of heart, much less the Best Buy mouth breathers.
AMD is doing something brilliant here, with the release of Catalyst 8.12 in early December, AMD is going to roll the CAL into the mainstream consumer driver. When you get the graphics drivers, you get GPGPU functionality for free, no download, no install... it just works.
What good is that without software, though? Not much, so that is being addressed as well. AMD defines GPGPU as just about everything not traditionally seen as a GPU task. This means anything other than 3D and video playback is GPGPU. Video encoding, game physics, and science are all classic examples, but you can't help thinking that the killer app is still in someone's basement, being built line by line evenings and weekends.
AMD has all of this, but they aren't being abjectly stupid about it like Nvidia. Over at the other green team, even the cafeteria food has a CUDA tag on it. Trust me, chocolate cake doesn't need to be made with CUDA, and there are some things that even bacon can't make better. AMD is putting things out, Nvidia is banging a drum, take your pick.
A good example of this is the video encoder. Nvidia has been hyping up Badaboom, a custom built app for video encoding. It works great as long as you need to go from one very specific format to another specific format, and quality is not a real concern. ATI's AVIVO encoder is rolled into the 8.12 driver pack, and it is free. Badaboom is $30. And it is slower.
The next gen of tools, SDK 1.3 will be out sometime in Q4, and given that the quarter is half over, this means pretty soon. It will work with all 4xxx series cards, and possibly older versions, but that is not a blanket statement. Stream will be in everything they make going forward, from integrated chipsets to 4870X2s.
Photoshop/CS4 acceleration is there. Silverlight acceleration is there, DRM and all. Office acceleration is there. Cyberlink and Arcsoft acceleration is there as well, and Arcsoft is said to have ported things in about four weeks. In fact, every mainstream app that Nvidia claims support for seems to be supported by Stream as well. Imagine that.
The main catalyst for this is OpenCL, the new GPGPU API that is being worked on by the Khronos group. Think of it as doing for GPGPU what OpenGL did for graphics. It also supports DXnext and OpenGL where applicable, so it is totally standards based. On top of that, you can write whatever you want, the Brook+ language is the one that AMD is pushing, along with several middleware layers, but you can use any OpenCL aware tool.
In the end, the difference is that Nvidia sees GPGPU as a way to drive margins and force people into higher priced cards. ATI is making it ubiquitous and free. Guess which one tends to win in markets, expensive and restricted or cheap and everywhere? If you are still confused, go look up this little company from Redmond called Microsoft...
Should you want the full OpenGL certified drivers for things like AutoCAD and Catia, you can get that too. The new FireStream 9270 is basically a HD4870 with 2GB of DDR5 and a very special driver set. It costs an eye-watering $1,499, but that is about $1,000 less than the competing Nvidia Quadro while packing twice the DP flops. If you need the piece of paper that says it will work with your software, this is your card. For everyone else, it is free. And included.
In the end, AMD/ATI is not being quiet any more. It has the fastest cards on the market, it has a much cheaper professional card, and now it has the APIs. In a few weeks, it will have the CAL everywhere, and what the competition charges an arm and a leg for, it gives out for free.
If Nvidia is counting on its professional line to buoy its bottom line, the next year is going to be ugly. ยต
Share this:
ShareBusiness Intelligence Software
Customer Relationship Management (CRM)
Enterprise Accounting Software
Enterprise Resource Planning (ERP)
Enterprise Systems Management
If Charlie is correct on all of this, I will shed a tear. It's my Computer Engineering degree's dream come true. Ubiquitous parallel processing. I'm so happy...
Looks like AMD is doing to nVidia what my dad told my wife to do, Grab, Twist and Pull

Ouch!
.. but now all you can see in GPGPU processing done on mainstrean VGA is CUDA. I have done some serious research and development in this. 

CUDA has one big advantage over Stream - Usable version is out on the market for about a year and many developers are quite familiar with its toolkit. An example of that can be seen at CUDA forum swhere you can find peretty good NV CUDA sanctuary :)

But I still wish for new open toolkit which will be supported by both NV/ATI and i don't like to be forced to write two different parts of code for one algorithm :(
How do you get Catalyst 8.12 onto your computer without a download and install?
Nvidia seems better with 32 to 48 Pips, Ati has 16. Ati has 3X redundancy, while Nvidia has straight shot. Yet, through ALL of reposts from ati1950 on, Little mention is made of vista Ultimate. whad'ya gonnaplay it.on?

7 won't work any easier for Gzmes than Ultimate. So Chuck, why Not Hurl about how Vantage do in 64 bit(Nessessary extra, these days) On These Units?.

Ati brings sub $300 price into focus, God, Entire computer isn't worth too much more. Pubes await GtX290, while lapping game pad up&up. 
Public also is sleeping on 800 based chipset for Ultimat. Cram it All in Box, ADD Ultimate64 & Game. Its Developing market, NT6 is Mandatory,NOW....

So Charles, Your Assignment, If You Accept it, Is Make NEW Cards Soundly(hdmi) Beat xp with Ultimate/7.Yeah, Soundly, That Reads Well.
dsl Drashek
That's another proof that ATi is the best.
Try to get this clarfied:
"It will work with all 4xxx series cards, and possibly older versions, but that is not a blanket statement. Stream will be in everything they make going forward, from integrated chipsets to 4870X2s.

Does that mean they do not plan to support current generation 780G (HD 3200) chipsets and that you have to wait for the 8XX series of chipsets for HTPC?
Thanks Charlie for this early-on information.

I have been wondering about when to buy the HD4670, but this news makes me want to buy it even sooner!

Cudo's to ATI/AMD they are really thinking about the customer!
I agree that amd has better products than nvidia right now, but, apart from that, I can't trust the rest of the article...

and not because I think charlie is helplessly biased, which he may or may not be: it's because those promises of widespread acceleration have been made before and not fulfilled

also, it's nice to see AMD/ATI trying to push mainstream cards instead of forcing users into buying the super-expensive pro versions, but AMD had better solve the many bugs in their firegl/firepro drivers if they want pro users to use their products (I know about the Maya bugs: they're awful, and have been there for over a year, untouched)
I, too, cannot wait to sniff around my 4870 :D

Access to such high performance hardware makes many a geeky software engineer weak at the knees! :D
to see, that my one year old combi of 690 mobo with integrated 1250 plus hd2400 might not get this enhencements

:-( 
grrrrrrhh
Once Again, Charlie D(Mr. anti-NVidia), great article.
AMD, I love this company... who doesn't? in the deep of your soul?

AMD's philosophy is so user oriented now... but they also know that when a company benefits the user, the company gets benefit back from the user. That's how it works for AMD. 
I would love to buy an ATI 4000. 
I would love to buy an AM3 CPU and use it on my AM2 motherboard. MSI and Asus just released a list of actual mobos that will be compatible with AM3 CPU.
4830 on XF? perfect!

And I probably would buy AMD, despite Intel has the greatest CPU now, but for a price.

Intel... that's another beast altogether
"If you are still confused, go look up this little company from Redmond called Microsoft..."

Did I almost hear a compliment to Microsoft on the INQ? Getting chilly around hell.

NJDevil.com
"there are some things that even bacon can't make better"

This is blasphemy and an affront to pig-loving folk everywhere! Shame on you Charlie for even making such an unfounded statement in an attemot at journalistic hyperbole... tsk, tsk,tsk...
" Guess which one tends to win in markets, expensive and restricted or cheap and everywhere? If you are still confused, go look up this little company from Redmond called Microsoft..."

So you're saying expensive and restricted wins in the market? How does this support your thesis exactly?? 

Perhaps you're harkening back to days of yore, old fella. At least one entire generation of humans has been borne since MS was considered cheap. Of course as fast as the years must go for you that was just yesterday, wasn't it?
I hope that ATI will also ship a Directshow-based H.264 decoder (non-DXVA) instead of just an encoder. Even nVidia does not have this one yet.
Seriously, can you try to be less objective about nVidia? GPGPU has been touted as the future for years, and ATI/AMD has been far behind, because they didn't dedicate actual hardware in silicon to it. CUDA is different from the past because it offers the graphics hardware as an alternative pipeline for general-purpose computing, so one doesn't have to know anything about graphics (e.g. OpenGL or Cg) to program it.

ATI/AMD keeps changing their mind about all of this. First, it was CTM (Close to the Metal), which required some assembler, and no one adopted it. Then came Brook+, based on what the CS department at Stanford did, and again, no one really caught on. OpenCL---which nVidia will support---doesn't yet exist, and when it does, it will reflect a lot of the features that are already in CUDA, simply because of market share.

When it comes to actually talking about real market share of GPGPU programs, the entire scene is about CUDA. No one I've ever heard of actually does real work with anything else. It's not because CUDA is perfect (far from it) but because AMD/ATI's solutions have NEVER caught on generally.

Writing that they're going to take over the market with this is just a total pile of crap. And to suggest that this has anything to do with their professional cards is even dumber. The code runs exactly the same on the GeForce GTX280 and the Tesla---and I should know, since I have both.

I generally like reading the Inq because it's generally a lot more interesting and honest than other sites, but this article about nVidia is just total garbage. You may have your reasons for hating the company, but next time try to look less like a total fool when writing blatantly incorrect and idiotic things.
What will the noticable differences be for users?

If I have a dual or quad core cpu and mid range gpu, what benefit will I see?

If I have a mid range CPU and low-end or on-motherboard GPU, what benefit will I see?

I was using some DELL pc's the other day that had on-board GPU, tiny cache on CPU and slow hard drives, with 256MB of RAM for XP. The customer thought they had bought a good pc, after all it was a DELL and that's good, right? Would this help them, the masses?
"AMD is doing something brilliant here, with the release of Catalyst 8.12 in early December, AMD is going to roll the CAL into the mainstream consumer driver. When you get the graphics drivers, you get GPGPU functionality for free, no download, no install... it just works."

What the heck is he talking about? CUDA is built into nVidia drivers by default. nVidia GPUs obliterate AMD GPUs in Folding@Home. Disproportionally so.

I still wonder who is giving the reach around in all this FUD, Charlie or AMD? Maybe it's a mutual circle jerk love fest.
played with Badaboom on rather low end 8500GT and it works fine. Clearly faster than using the CPU (Q6600). Read some reviews about the AMD Avivo, and it seems that there are some artifacts and high CPU useage.
We and the people haven't seen stream everywhere. And all we have heard of is CUDA everywhere. NVIDIA CUDA is known as graphics plus. It's more than just graphics. This has been taught at over 100 Universities now. I'd like to see ATI puts stream everywhere like what was mentioned in the article, but this is the next step. Now I think ATI/AMD has trouble of competing with Nvidia CUDA. Eventhough they are able to put stream out there in the market, it will be very hard to convince people and turn the market around due to the efficiency and popularity of CUDA. I am using CUDA on my labtop and workstation. It's amazing.