Back from the Dead! (Part 2)

Another long hiatus, another long post depression. It looks like the World Economies got wind of my decrease in posting and suddenly we are spiraling toward a worldwide depression! 
Meanwhile, one of my favorite processor companies gets into hot water and splits into two

And now I learn that even my laptop will inevitably fall victim to a faulty GPU, exploding into flames and taking out my desk along with it! Oh how will I play Crysis Warhead now!?! Good thing the repair should be free.

Now guess what, due to a recently discovered vulernability in adobe flash, all browsers are capable of being “clickjacked” by nefarious persons. What does this mean? Click on the wrong link, and your microphone and webcam were just secretly activated by some creepy dudes in Eastern Europe. A fix is in the works, but until then, get used to the idea of being watched. 

Hydra Engine – Decimating SLI and CrossfireX

This article may go a little further beyond the normal tech scope of this blog, but the release today of over 8 articles by pro tech sites on the subject (which I had already done work researching and posting about) has convinced me to repost my original work and additional information here.
SLI and CrossfireX are Nvidia and  ATIs respective technologies for combining multiple Graphics Cards in a single computer system. While most of us get along just fine (or otherwise) with a single GPU, enthusiasts have the option of utilizing two or three (or four) GPUs to supercharge their gaming performance. Scientists have also discovered that these multi-GPU setups can greatly benefit compute intensive research applications such as Folding@Home. ATI entirely replaced their high-end GPUs with their two high performance cards fused onto a single circuit board, creating the impressive 4870X2 (and previously the 3870X2) as a result.

Hydra Engine

All of this technology is run by either SLI and Crossfire, technologies that attempt to share video rendering load over several GPUs by having each GPU render every other frame or half (or a third) of each frame. Problem is, in most cases performance scaling is not linear. In other words, two graphics cards don’t give you twice as much performance, maybe 70% extra at best. A third and fourth GPU may only increase performance 10% and then 5%, in many cases. Getting games to scale properly is extremely hard work for the developer and coders responsible for SLI and CrossFireX functionality. It is very hard to justify 3 GPUs getting you 50 FPS for a total cost of 900 dollars when a single GPU will get you 30 FPS. Additionally, the cards are using 600 watts when you could be using 200 much more efficiently. While SLI and CrossfireX have been slowly improving their scaling, they are nowhere near perfect, and they often have side-effects. The performance benefits only exist if a system is running a game full-screen,  and it is impossible to run two screens while utilizing the technologies. VERY recent developments may have begun to alleviate these issues, but they have been a long time waiting.

It seems a company called LucidLogix may beat Nvidia and ATI at their own game. PC Perspective has written an article detailing how Lucid’s hardware and software may allow for perfect GPU scaling using an unlimited number of GPUs of any model within a respective brand.

Here is a small clip of the two page article:

What is the HYDRA Engine?

At its most basic level the HYDRA Engine is an attempt to build a completely GPU-independent graphics scaling technology – imagine having NVIDIA graphics cards from the GeForce 6600 to the GTX 280 working together with little to no software overhead with nearly linear performance scaling.  HYDRA uses both software and hardware designed by Lucid to improve gaming performance seamlessly to the application and graphics cards themselves and uses dedicated hardware logic to balance graphics information between the CPU and GPUs.

Why does Lucid feel the traditional methods that NVIDIA and AMD/ATI have been implementing are not up to the challenge?  The two primary multi-GPU rendering modes that both companies use are split frame rendering and alternate frame rendering.  Lucid challenges that both have significant pitfalls that their HYDRA Engine technology can correct.  For split frame rendering the down side is the need for all GPUs to replicate ALL the texture and geometry data and thus memory bandwidth and geometry shader limitations of a single GPU remain.  For alternate frame rendering the drawback is latency introduced by alternating frames between X GPUs and latency required for inter-frame dependency resolution.

Link to full article

Answering a question:

Harleyquin: Interesting concept, but how does this translate into improved gaming performance on multiple GPUs?

They claim almost perfectly linear performance.

(all examples use made up starting FPS values)

Simple example: 3 x ATI 4850

1 x 4850 = 15 FPS in Crysis at Max Settings.
2 x 4850 = 30 FPS in Crysis at Max Settings.
3 x 4850 = 45 FPS in Crysis at Max Settings.

More complicated example: 1 x Nvidia GTX 280, 1 x Nvidia 9800GTX, 1 x Nvidia 8800GT

GTX 280 = 25 FPS in Crysis at Max Settings.
9800 GTX = 20 FPS in Crysis at Max Settings.
8800 GT = 10 FPS in Crysis at Max Settings.

GTX 280 + 9800 GTX = 45 FPS in Crysis at Max Settings.
GTX 280 + 9800 GTX + 8800 GT = 55 FPS in Crysis at Max Settings.

So essentially, Hydra is taking the place of CrossfireX and SLI, and according to their claims they make perfect use of what they are given, no wasted GPU power.

They claim we’ll see this integrated into motherboards and certain GPU boards by 2009. In other words, any motherboard with this chip will be able to run either multiple ATI cards or multiple Nvidia cards without special licensing by either company. This really affects Nvidia and SLI more strongly, as ATI is already pretty lose with CrossfireX licensing. This could do to gaming what dual and quad core processors did to single core computing. Cheaper, faster, and more efficient use of power. Sounds good, no?

Integrated Graphics, AMD’s PUMA

IGP (Integrated Graphics Platforms) have been the bane of gamers, especially mobile gamers, for many years. Created for use by casual computer users who want little more than to write Word documents, play flash games, and watch YouTube, they often fall well below the mark for acceptable DVD/Blueray playback and PC Gaming. For mobile users, the choice between a dedicated GPU (Graphical Processing Unit) and an Integrated GPU merits consideration. Integrated GPUs use much less power, create less heat, and result in an all around more mobile system. Dedicated GPUs use more power, create much more heat, but allow gamers on the go to play the latest and greatest with a few tweaked settings. Intel’s IGP solutions, the GMA series of integrated graphics (GMA 915, GMA 950, X3100, upcoming X4500), have been the most common and also the most frustrating. Intel doesn’t do graphics, so their IGPs are very hit and miss, especially in running games properly. ATI/AMD’s solutions, such as the X1250, as well as Nvidia’s solutions, such as the 7150, provide more acceptable gaming performance, but still not enough to play high intensity games very effectively at medium/high settings.
AMD/ATI finally broke the mold, their new 780G motherboard chipset contains the HD3200 IGP, which offers a significant performance improvement over all other IGPs. While a dedicated GPU will still perform better, the HD3200 is an important step forward. In a mobile system, the 780G is part of AMD’s PUMA platform, which also offers an external port allowing an external GPU to be connected to the laptop, something not practical or effective until this point. The following video compares the HD3200 with an older X3100 (which has been replaced by the slightly improved, though still lacking, X4500).

AMD’s 4850

Recently I wrote about Nvidia’s new GPU series, the GTX 260/280 cards. These beasts cut a swath through the current video cards, handily stomping anything that had come before, but they did so at a high price, $400 and $650, respectively. Gamers appreciate these cards, but often it is hard for anyone to justify so much money on a single component when an entire computer can be purchased for the same amount.
Radeon HD 4850

Now, various sites have released benchmarks of the AMD 4850, and it has me happily surprised.

I spent a large enough word count getting into GPU specifics last time, so I’ll just outline the major points of this card. First, it performs very much like an Nvidia 9800GTX, which (before today) ran a hefty $300. That should stop most people right there, a 200 dollar card offering the same (or better) performance as a 300 dollar card. Well, in an attempt to rain on AMD’s parade, Nvidia has drastically cut the price of the 9800GTX down to $200! So really, considering the prices, the real fight is between the AMD 4850 and the Nvidia 9800GTX. They trade blows in most benchmarks, in some Nvidia pulls ahead and in others AMD is the champ, but in most cases performance differs by a small margin (10% or less). That said, the 4850 runs slightly cooler and requires slightly less wattage than the 9800GTX. It also only requires 1 six pin power supply connector, while the 9800GTX requires 2.

Crysis DX10To the right, you can get a taste of how well the 4850 performs on the GPU-eating game we call Crysis (at various resolutions). Take a look at how well the 4850 performs not only against the 9800GTX but also against its fairly successful predecessor, the 3870.

Soon the 4870 will be released, offering MUCH faster memory and higher clock speeds. Some (highly excitable) reviewers have estimated as much as a 40% performance increase compared to the 4850. Soon after that will come the 4870X2, fusing two 4850 cores into a single piece of PCB board, and possibly beating the GTX280 to a pulp. Still, the price wont be quite as sweet, most estimations of the 4870 price range between $250 and $300. This will make for an interesting show in the coming months.

AMD/ATI has really blown the lid off this generation’s GPU battle. They have a competitive GPU at a competitive price with competitive features, and it is smaller, cooler, and slightly quieter and more power efficient than the 9800GTX. We haven’t seen this level of competition for almost 2 years now, and I’m glad it has returned.

Nvidia GTX 280/260 Benchmarks Released!

In advance of the launch tomorrow, any tech site worth its salt has released a hands-on review of Nvidia’s next line of GPUs. So far the nomenclature of the new series completely resets Nvidia’s old system. Previously, the last Nvidia line of GPUs ended (presumably) at the 9800GX2. Now they are back in the hundreds and have placed a GTX prefix in front of the entire line (so far).
Ok, on to the random bits of information. This beast has 1.4 BILLION transistors. Nvidia’s last high-end card had only 754 million transistors, so we are seeing almost double the brute force capability. Let’s show you a little comparison between a Geforce GTX280 and a top of the line dual core Penryn CPU from Intel:

GTX 280 Die Size

Now technically, in a simplified sense, the older generation of Nvidia GPUs had 128 cores. Consumer CPUs, at most, have 4 cores. The new generation of Nvidia GPUs has 240 cores. That is an INSANE number of cores compared to a Central Processing Unit. Maybe that’s why the GTX 280 can suck 236 Watts at full bore. Now you might wonder, why not replace or augment your CPU with all of that crazy power in your GPU? Well, the industry is actually moving in that direction. The primary roadblock is the fact that GPUs process data in a very specific, specialized way and CPUs are built to process data in a very general way. GPUs are generally built to work with pixels, while CPUs are built to work with ANY data. We’ve already seen GPUs used by scientists to do brute force calculations much faster than CPUs, and we’ll see a more mainstream/consumer fusion of the two components in late 2009.

So how much faster is it?! Compared to the previous single-card solution from Nvidia, the 9800GTX, it is roughly 50% faster in most games. Running the game Oblivion, here is a benchmark graph stolen from comparing the top performing GPUs at maximum resolution and maximum detail. This resolution is so high that it is confined to 30 inch monitors, most users will not be pushing their GPU nearly this hard. Score is presented as number of rendered ‘frames’ per second.

GTX 280 Benchmarks

The GTX280 will cost $650 and the GTX260 will cost $400.

GTX 280

Nvidia has done it again, the fastest single GPU in existence. AMD/ATI have their work cut out for them, we already know their 4870 wont be as fast as the GTX280, but theoretically the pricing should be MUCH lower and multi-GPU solutions could end up giving them a competitive advantage performance-wise. We’re in the midst of another GPU war! It’ll be a few more days till we get to see direct comparisons between the best from AMD and the best from Nvidia.

I. Can’t. Wait.

Week of 6/15/08

We’ve got three events upcoming!
On the 17th:

Firefox 3 will be released. Everyone go out and get it on launch day, we are trying to set a world record for most downloads in a single day.


The Spore Creature Creator will be released, more on this Tuesday.


On the 18th:

Nvidia releases their next line of GPUs! It seems only a few months back the 9000 series hit the shelves, now the GTX 200 line is already at our doorstep. AMD will be releasing their 4000 series early next week in retaliation.