Slashdot Mirror
Intel Abandons Discrete Graphics
Stoobalou writes with this excerpt from Thinq: "Paul Otellini may think there's still life in Intel's Larrabee discrete graphics project, but the other guys at Intel don't appear to share his optimism. Intel's director of product and technology media relations, Bill Kircos, has just written a blog about Intel's graphics strategy, revealing that any plans for a discrete graphics card have been shelved for at least the foreseeable future. 'We will not bring a discrete graphics product to market,' stated Kircos, 'at least in the short-term.' He added that Intel had 'missed some key product milestones' in the development of the discrete Larrabee product, and said that the company's graphics division is now 'focused on processor graphics.'"
I hope they at least manage to incorporate some of what they've learnt into their integrated chips.
Intel's integrated chips have been appallingly bad in the past, some incapable of decoding HD video with reasonable performance. Manufacturers using those intel integrated chips in their consumer level computers did a great deal of harm to the computer games industry.
They've never been able to bring the most innovative designs to market.. they bring 'good enough' wrapped in the x86 instruction set.
If x86 was available to all I think we'd see Intel regress to a foundry business model.
' He added that Intel had 'missed some key product milestones' in the development of the discrete Larrabee product,
Like proof that they were even capable of making an integrated graphics product that wasn't a pile of garbage?
GMA910: Couldn't run WDDM, thus couldn't run Aero, central to the "Vista capable" Lawsuits
GMA500: decent hardware, crappy drivers under Windows, virtually non-existant Linux drivers, worse performance than GMA950 in Netbooks.
Pressure to lockout competing video chipsets. We're lucky ION saw the light of day. http://www.pcgameshardware.com/aid,680035/Nvidia-versus-Intel-Nvidia-files-lawsuit-against-Intel/News/
Short of buying out Nvidia I don't see Intel having a consumer's chance in America of competing with AMD in the value sector for the next few generations of chips.
CPUs have been "fast enough" for years, but GPUs have not. AMD is going to laugh all the way to the bank being able to offer a $50 package that can run The Sims.
ignoring you're complete inability to form a sentence
Hey everybody, 'tard fight! Come watch!
This is bad news for one reason. Competition. There are only 2 major players in discreet graphics right now and that is horrible for the consumer. Now the good. Intel SUCKS at making gpus. I mean seriously. So either way Intel has no hope of making a 120 core GPU based off of x86 being cheap or fast enough to compete. Go big or stay at home. Intel stay at home.
A large, publicly announced project with a great deal of media hype that had the potential to shake up the industry was cancelled. So, yeah, stop the presses.
Everyone gets up on Intel integrated GPUs because they are slow, but they are looking at it from a gamer perspective. Yes, they suck ass for games, however that is NOT what they are for. Their intended purpose is to be cheap solutions for basic video, including things like Aero. This they do quite well. A modern Intel GMA does a fine job of this. They are also extremely low power, especially new newest ones that you find right on the Core i5 line in laptops.
Now what AMD may do well in is a budget gaming market. Perhaps they will roll out solutions that cost less than a discreet graphics card, but perform better than a GMA for games. That may be a market they could do well in. However they aren't going to "kill" Intel by any stretch of the imagination. For low power, non-gaming stuff using minimal power is the key and the GMA chips are great at that. For the majority of gaming, a discreet solution isn't a problem ($100 gets you a very nice gaming card these days) and can be upgraded.
The i740 card.... great expections, poor real world experience.
its potatoe you dumb fuck.
More directly, what the hell is "discrete graphics"?
It refers to a graphics processor as a separate (discrete) component of a computer system. A chip that does nothing but graphics can be more powerful than integrated graphics because the GPU circuitry doesn't have to share a die with the rest of the northbridge.
To be fair to Intel, most graphics cards then were on the PCI bus, not AGP, so they didn't have the opportunity to use the host RAM except via a very slow mechanism. At the time, the amount of RAM was far more of a limitation than the speed, and a card using 8MB of host RAM via AGP was likely to have an advantage over a card with 4MB of local RAM on the PCI bus. While it was much slower than competing solutions, it was also much cheaper. The RAM on something like the VooDoo 2 was a significant proportion of the cost. A 740 cost about 20% of a VooDoo 2 and using system RAM had the advantage that you didn't have a load of RAM doing nothing while you were not doing 3D stuff. At the time the 740 was introduced, I had an 8MB VooDoo 2 and only 32MB of main memory. Having 8MB of RAM sitting doing nothing during the 90% of the time that I wasn't playing 3D games was a massive waste.
I am TheRaven on Soylent News
It's not a new term, and it's not unique to GPUs. The distinction between integrated and discrete coprocessors has been around for at least 25 years. If you read something like Byte from the early '90s, you will find discussions about the relative merits of integrated and discrete FPUs. You'll find a similar discussion on integrated and discrete MMUs and various other components if you look a few years earlier.
I am TheRaven on Soylent News
Actually Intel have changed the name to NotToBee.
I kind of think Larrabee was a hedge.
If you think about it, around the time it was announced (very early on in development, which is not normal), you had a bunch of potentially scary things going on in the market.
Cell came out with a potentially disruptive design, Nvidia was gaining ground in the HPC market, OpenCL was being brought forth by Apple to request a standard in hybrid computing.
All of sudden it looked like maybe Intel was a little too far behind.
Solution: Announce a new design of their own to crush the competition! In Intel-land, sometimes the announcement is as big as the GA. Heck, the announcement of Itanium was enough to kill off a few architectures. They would announce Larrabee as a discrete graphics chip to get gamers to subsidize development and....profit!
Lucky for them, Cell never found a big enough market and Nvidia had a few missteps of their own. Also, Nehalem turned out to be successful. Add all that up, and it becomes kind of clear that Larrebee was no longer needed, negating the fact that it was a huge failure, performance-wise.
Intel is the only company that can afford such huge hedge bets. Looks like maybe another one is coming to attack the ARM threat. We'll see.
FUNK!
Note for anyone else whose curiosity was piqued, this only works with 32bit systems with 950 chipset based systems, and does not work with GMA X3100, GMA X4500, GMA 500, or GMA 900.
The Larrabee chips actually looked pretty good. There was a lot of hype, especially from Intel. They demoed things like Quake Wars running a custom real-time ray-tracing renderer at a pretty decent resolution. Being able to use even a partial x86 ISA for shaders would have been a massive improvement as well, both in capabilities and performance.
From what I've been able to piece together, the problem wasn't even the hardware, it was the drivers. Apparently, writing what amounts to a software renderer for OpenGL/DirectX that got good performance was beyond them.
Another part was an odd insistence on doing all the rendering in software, even stuff like texel lookup and blitting, but that's another story.
From what I remember all the cards I was using at the time Intel was trying to sell the i740 (Permedia-2, TNT, etc) were on the AGP bus.
Check the dates. The i740 was one of the very first cards to use AGP. Not sure about the Permedia-2, but the TNT was introduced six months after the i740 and cost significantly more (about four times as much, as I recall). It performed a lot better, but that wasn't really surprising.
I am TheRaven on Soylent News
If you read something like Byte from the early '90s, you will find discussions about the relative merits of integrated and discrete FPUs.
yikes, my memory of installing an 80387 has been completely un-accessed for at least a decade. Thanks for the scrub. :)
My God, it's Full of Source!
OUTSIDE_IP=$(dig +short my.ip @outsideip.net)
There was a company called Rapid Mind, which built library & tools for writing code to target various GPUs, multi-core CPUs, etc. Something similar OpenCL, I suppose, but easier to program (theoretically -- I never actually tried it). Intel bought it and killed it.
Another company, Havok, developed a successful physics & AI library. They were going to port it to to GPUs. Then Intel bought it and canceled the GPU port.
___
If you think big enough, you'll never have to do it.
"Graphics cards with performance comparable to the best integrated graphics aren't exactly expensive" ...).
You can't find expansion graphic cards with performance comparable to the current integrated graphics - the integrated graphics are slower than anything else (less available memory bandwidth, fewer compute clusters,