Tuesday, December 11, 2007

As I've thought more about Larrabee, I have some additional observations:
I'd like to clarify my point about the x86 instruction set. It's not innately bad, but it doesn't progress the architecture. The x86 ISA has too few registers, an irregular instruction length, and highly irregular instruction decoding - none of which help with architectural efficiency. It's just all the work and R&D put into software for the x86 architecture that compensates for its shortcomings. But if given a clean slate, few would chose the x86 ISA for maximum performance or efficiency. My point is that x86 is largely irrelevant for a GPU. The x86 will help Intel in the HPC market.

With regard to the wide SIMD (SSE) structure - we'll have to see how effective Intel is in packing useful operations. It seems that the width was chosen in order to hit certain theoretical performance metrics, not for realistic workloads. The same ambition infected the Cell processor and we're still waiting for software to approach those highly promoted peak numbers.

I remain convinced that Intel doesn't have the formula to produce a world class GPU. I see no sign they have the specialized talents to make it succeed. At best, it will be a reasonably sized die with mediocre performance, but it will have Intel's brand name behind it. That will sell some units, but I hope Intel is ready for the realities of GPU ASPs in the mainstream of the market. At least it should be better than the existing Extreme (-ly bad) Graphics they have today.

I also realized that most PC buyers have no idea how bad Intel's integrated graphics really is. The gamer sites never test Intel graphics because they know how bad it is and it's not even worth the effort. But the mainstream consumers don't know how bad it is because NO ONE TELLS THEM! Both ATI and NVIDIA are guilty here because we still have to work with the Intel chipset group and neither company has been willing to take Intel on directly. Shame on us. It's also not a topic that consumer publications make a big point about, as many in the industry consider it common knowledge - its not new news to them. So Intel gets a free ride. I was waiting for AMD to push it's platform strategy and call out Intel graphics, but I haven't seem it yet. Maybe when its new DX10 chipset ships in Q1.

With regard to Larrabee's supposed process advantage, I find it difficult to image Intel will use the leading edge process for Larrabee. I expect the chip and its successors will be 6 months to a year behind the leading edge of the process. If that's the case, then TSMC will be very competitive. In addition, our design process is great at turning around designs quickly, Intel's process is geared more toward die optimizations with its custom circuit design. Intel will likely make Larrabee coherent so that multi-chip scaling will allow Intel to make the one die and still create multiple graphics card solutions. We tend toward multiple design solutions with one die optimized for a specific price point. AMD may also be moving to the same idea as Intel. We'll have to follow that development closely.

So, if anyone else would like to comment, please add your voice. I do moderate the comments, but I promise to publish any reasonable response.

Friday, December 07, 2007

My last post did generate a response, so check it out below and my response to the comment.

I should add that I was talking from my perspective. NVIDIA does take the Larrabee program very seriously. I see it more as a research project gone amok.

What I also find amusing is that Intel, for many years, tried to run away from the x86 instruction set with i960, i860, i432, EPIC (Itanium), and xScale, and now x86 is the answer to all applications. Fred Weber when he was CTO of AMD was the first I know to active promote the "x86 everywhere" strategy. While AMD still embraces this notion, I haven't seen them approach graphics directly with the x86 instruction set.
I think that AMD's Fusion is a wiser move overall.

Wednesday, December 05, 2007

It's a bit late (actually you may note I tend to blog late at night anyway), but I've been wanting to take on a topic that's of interest to me from both an NVIDIA point of view and my Microprocessor Report analyst background. It's the chances that Intel's Larrabee will be a successful GPU.

I would like to offer the opinion that the deck is stacked against it, despite Intel's hype machine pushing it. For one, the design looks more like the IBM Cell processor or the Sun Niagara (UltraSPARC T2) than a graphics chips. The Cell was proved to be ineffectual as a GPU (which explains why we got the graphics chip in the PS3) and Niagara is focused on Web services (lots of threads to service).

Let me also ask this question: when has Intel EVER produced a high performance graphic chip? (Answer: never)

Oh, and when did we all decide that x86 was the most perfect instruction set for graphics? Hmmmm, it's not. The example where the x86 is the preferred instruction set for either graphics or high performance computing doesn't exist. The x86 instruction is an burden on the chip, not an advantage, for graphics. In fact, graphics instruction sets are hidden by the OpenGL and DirectX API, so the only people who care are the driver programmers and we are free to design whatever instruction set is most efficient for the task, while Intel has handcuffed themselves with a fixed instruction set. Not the smartest move for graphics. Intel's plans may work better for HPC though. Still, there are many alternatives to x86 that are actually better (shocking isn't it)!

Larrabee is based on a simplified x86 core with an extended SSE-like SIMD processing unit. In order to be efficient, the extended-SSE unit needs to be packed efficiently. How is Intel going to do that? The magic is all left to the software compiler, which is going have a tough time finding that much parallelism efficiently.

Larrabee will have impressive theoretical specs that will wow the press, but it's going to be very hard to use the chip up to its potential, which sounds exactly like Cell. And Cell has not lived up to its hype. So my corollary is Larrabee = Cell; Cell not = to hype; Larrabee will not = hype.

Now we wait until sometime in mid-2008 to see some silicon. But you heard it hear first: Larrabee will not reach the potential performance capability (by a long shot) and will not displace NVIDIA as GPU leader.
Sorry for the lack of postings but I'm now been caught up in the game Crysis. Excellent gaming experience. I like.

Monday, December 03, 2007

Hi Ho Everybody, I'm back. I haven't had a lot of time with the Eee PC, but it's still reasonably nice. I'll like it a lot better when ASUS increases the screen resolution. It would also be nice to move up to a Core 2 processor with real power management, but these are the trade-offs you make when its only $400.

I've been in the middle of thinking about AMD, Intel, and NVIDIA as we all do this dance around the ring - its like a three-way fight where we take turns helping and hurting each other. I'd call us the Good, the Bad, and the Ugly (I'd let you all decide who is who, but I know my choices).

On the topic of ASUS, I'm beginning to wonder about the quality of their motherboards. Both my stepson and I have problems with the motherboards not booting properly and these are two completely different motherboards. How hard can that be to get right?

One project I recently completed was taking video at the memorial for the mother of my wife's brother in-law. I used Pinnacle Studio 10.5 and found it pretty easy to use and create a good basic video with fades and titles. I want to do more work with it over the holidays.

Wednesday, November 14, 2007

There's an extensive review and guide to hacking the Eee PC on Ars Technica.
There was also a discussion on TWiT 121 with Patrick Norton being the skeptic and Wil Wheaton Harris willing to give it a chance. I think it's the most interesting (new) computer $400 can buy.

I haven't had the opportunity to explore it any further, I went to the SIA dinner with John Cleese as the guest comic. Rather funny guy :-)

Tuesday, November 13, 2007

New gadget in the house! I bought an ASUS ~$400 Eee PC. I was fighting the urge to buy it for two weeks now and finally it won. I could have waited for the 8GB model or the black model, but in a moment of weakness, I succumbed.

I've go the wireless set up for home and started to try out the various features. The mike is reasonable, the video is pretty crappy, the speakers are weak but sufficient, the keyboard is small and easy to hit too many keys (my typing sucks anyway), and the 7 inch display (800x480) is very small but it is bright enough for indoors (haven't tried it in daylight yet).

We'll see how it goes, but so far I like it despite its shortcomings.

Tuesday, November 06, 2007

My response to TWiT 120

First, I told Leo that I couldn't make the show for two reasons: one, was that I had a friend from Chicago coming into town; and, two, was that I didn't think there was anything I could provide an interesting comment on as the preceding week's news was more about software/Leopard and Internet business related.

As it turns out, I was wrong on both counts. My friend and I didn't connect on Sunday afterall and this TWiT ran into some interesting questions on phones and PCs. I should have joined, but then I would have missed the Patriots and Colts game, and as much as I am a techie, I still wanted to watch that game. Yes, I could have recorded it and watched it later (fast-forwarding through the commercials), but I just don't enjoy football that way. While I was watching the game, I also found time to move my game PC into a new case with a new power supply. The case is the Antec 900 gaming case which was on sale at Frys for 1/2 price, too good to pass up. But I digress.

Despite not being there, my name did come up with regards to NVIDIA's participation in the Google's Open Handset Alliance (Android) project. My decision to not participate in TWiT 120 had nothing to do with this announcement. Google's announcement was not a big deal for us as we don't have a lot of handset sales right now. And, the announcement was all about the software stack, not about the hardware, so I stand by my statement to Leo that there wasn't much hardware news.

Where I would have liked to interject were on the discussion on the PC dying as a platform. Dvorak would have also been a good foil for this debate. The most dynamic, transformational platform is still the PC (and I include the Mac here). A good example is the new Asus Eee PC. It's really small, and not real fast, but its only $400, runs a robust and lean Linux desktop, its very light, connected, and portable. And it's selling like Wiis. The PC will continue to evolve and reach new markets. I still think it's the best gaming device (well maybe that because I never mastered the game controller). I won't deny that many millions of people will access the Internet first through their cell phones, but I believe they will still prefer the PC (as primary Internet access device) if given the choice.

Thursday, November 01, 2007

Here's the update on the Vista game PC. I ditched Microsoft technical support and decided to start over again. This time a bought a 10,000RPM boot drive and installed the O/S on that drive, saving the 500G RAID 0 drive for applications (games). I first installed the OEM copy of Win XP MCE I had originally build the system with. I didn't bother to install any of the service packs or even authenticate it, in fact I disconnected it from the Ethernet to make sure it didn't phone home or get infected. Then I immediately installed my Vista Premium upgrade edition. Everything went smoothly. I then installed all the latest drivers. And it's doing very well. The only problem I've had has been standby issues with the ASUS motherboard and Vista. Occasionally the motherboard hangs in an ACPI state. I've also noted that the motherboard occasionally won't cold boot until I reset it or power cycle it. That is definitely a motherboard issue.

Now I've got some of the best new games installed: Bioshock, Valve's Orange Box, World in Conflict, UT3 demo, and the Crysis demo. My favorite "old" game, Battlefield 2, won't install though. I need to figure that one out.

Tuesday, October 23, 2007

So I spent two and a half hours with the Microsoft support line and still no resolution to the problem. I've concluded that I will have to re-install WinXP and then re-install Vista. But first, I'm going to install a new 10,000 RPM main drive and re-install the O/S on that drive. Using a RAID 0 drive as the boot drive caused way too many problems with the WinXP install.

On a side note, I picked up 2001: A Space Odyssey on HD DVD. Early informal reviews at AVSForums has been positive. The old DVD I had was horrible. I'm still trying to buy the Valve Orange Box at Fry's for the discount price (I have a rain check), but it's constantly out of stock. I wonder if Valve is shorting retail in order to drive more people to the Steam online delivery system, bypassing the retail channel and manufacturing costs. Naw, Valve wouldn't be that manipulative.

Sunday, October 21, 2007

Some quick blogging while I'm working with Microsoft tech services to re-enable my Vista. Fun.
We're at 1 hour 10 minutes and counting....

A little background: I had some instability with the rebuilt game PC system and I thought it would be better to upgrade to Vista with a new set of drivers. The Motherboard is an ASUS P5N-E SLI with 2GB of DRAM. It turns out the motherboard has had some DRAM stability issues that works best with a manual override. The system instability caused Vista to become corrupted, but the retail upgrade edition of Vista wouldn't let me reinstall it easily.

Now are 1 hour and 40 minutes and still no working Vista. The India-based tech has already complimented me twice on my patience.

The reason I wanted Vista on my game PC was for the DX10 content, with World in Conflict sitting on my shelf and Crysis coming. I may have to resort to reinstalling WinXP and trying to install Vista again. Cripes.

Sunday, October 07, 2007

I just finished my participation on TWiT 116! I'm almost an occasional semi-regular guest as this is my third TWiT (not counting the MacWorld episode where I was in the audience and also asked a question).

Not to sound too fawning, but I really enjoy the TWiT pod/netcasts. I usually listen to them on the way to work in the car.

If you'd like to make a comment about my guest appearance, please be kind. I'm more a print guy, than a radio guy, although I was a big fan of Bob and Ray. In college, on occasion, I would help a friend of mine on his college radio show with a sort of Bob and Ray-style improvisation. Bob and Ray humor is very dry, and is kind of a cross between Bob Newhart and Monty Python.

Unless Leo edits it out, at one point you'll hear a door slam and a rattling sound - that's the door to the garage next to my office closing (it's spring loaded to force it to close), which cause the metal shelve holding my train collection to vibrate (see pic).

One of the new products I'm going to review on this site is the Belkin N1 Vision Wireless Router. It's looks and functionality were too good to pass up.
I'm back!
After a fling with Vox, I'm back to Blogger and my EntroPC blog for technology. I'll keep Vox around for personal blogging about stuff like wine, beer, cars, etc. EntroPC will continue my technology-related thoughts and experiences.

For those who are wondering about the name, EntroPC is a pun on entropy and PCs. While I originally thought about the PC industry losing it's energy, I actually think that the industry has never been so dynamic. It keeps getting better and adapts to new markets. While I would argue that Intel's plans to move x86 processors into smartphones is not the most logical idea I ever heard, the development of small PC platforms (MIDs, UMPCs, OLPC) does show that the PC is very adaptable.

I also believe that the PC is the best gaming platform; better than consoles. The reason is that a PC can be adapted to to be an excellent game console, but a game console will never be a good PC. I'd love to see someone develop a game PC that did just that. I think Dell or HP could do it. It just seems that no one has the will to do it. The company I work for, NVIDIA, could do it, but it lacks the consumer channels, very deep packets, and will to compete with its own customers (Microsoft and Sony).