The WorkShop

The guys at The Enquirer seem to have some dough about the next generation GPU from  nVidia , In an analyst webcast, Nvidia Investor Relations and Communications VP Michael Hara stated the top end card based on the G92 graphics processor will be ready for Christmas and that it will have computing power close to 1 TeraFLOP .
Time for some cools stats now , Intel’s latest quad core Core 2 extreme QX6700 runs at 2.6 GHz and has a peak floating point performance of 50 GFLOPS, while nVidia’s G80 ( powers the 8800GTX)  has a peak floating point performance of 330 GFLOPS, AMD’s R600 can do a max 450 GFLOPS  and the STI Cell B.E does close to 250 GFLOPS at max throughput . So if the G92 rumor holds it would mean that the G92 will outperform the cell by around 4 times and its predecessor G80 by about 3 times  in terms of computing power.
One problem with GPU’s  is the inability to support 64 bit floating point operations, which is necessary for almost all supercomputing applications . On the other hand GPU’s are far cheaper than supercomputer vector processors making their use in HPC( High performance computing) justified . nVidia had promised FP64 support on GPUs by late 2007 , So is G92 the promised chip , if so its going to have a big impact on the GPGPU movement and on HPC in general.
One little concern for me is the lack of titles for the PC which can take advantage of such horsepower , Crysis is a contender and so is SC-Conviction . TeraFLOP or not it’ll be interesting to see what kind of performance can the G92 can dish out.

First the big news , BT and Sony computer entertainment Europe(SCEE) have signed a four year deal to transform the PSP by adding wireless broadband communications functions including high quality video and voice calls and instant messaging. At first, only PSP-to-PSP calls will be supported but this will be soon followed by the ability to support calls and messages between PSPs, computers, regular phones and mobiles. The service will first be introduced in the UK and the rest of Europe . Surprisingly no details were released on when the service will be available in the US which has a PSP user base of 7.4 million . I guess this is Sony’s answer to the Nintendo DS , which is set to break all sales records for a gaming hand held before its time runs out , but there are a hundred reasons why this strategy won’t work at least with the current avatar of the PSP . Here are my top 5 reasons

• Anybody who’s seen or held the PSP knows how big it really is, with its large screen and dedicated gaming buttons , its not something you would want to be seen talking into.
• There are already so many complaints about PSP’s battery life and with video and voice calls , its anybody’s guess how long the battery would last.
• Video calls will need a cam , and with Wi-fi the new hardware coming into the PSP will have some implications on the size of the device , will it get bulkier or is Sony going to rethink their whole design strategy for the PSP and release it as a new model with the current PSP existing as a low price model.
• The D pad and associated buttons cant be used for messaging , of course they know that , does that mean more buttons or or rather a small alphanumeric keypad . I think touchscreen would be better , what impact will that have on the cost and size of the next gen PSP.
• Last but not the least, the cost. Just how much is this gaming device cum phone cum Wi-Fi internet device cum media player going to cost , not as much as the PS3 I hope

What could be done otherwise is to improve the PSP as what it was meant to be , a portable gaming and media device : improve the battery life , provide descent internal storage, better content delivery and for god’s sake drop that damn UMD drive.

From the recent announcements of the chip majors , it seems the catch up games that they have been playing with each other are over.
What am I talking about ? First it was all about pushing clock speeds up , and more recently it’s been about putting more cores on a chip more efficiently than the other . But as the multicore architectures become more mature , chip majors like Intel and AMD have started to diverge on their multicore strategies . Intel is moving towards homogeneous cores while AMD and IBM look to be adopting a more heterogeneous approach .

Homogeneous as the word implies means all the processing units (cores) on a certain processor are of the same type and divide the workload between them for maximum efficiency . The heterogeneous approach is a little more complicated as it may be implemented in a lot of ways , different size cores for different dedicated functions is one approach . Another approach is where a problem’s workload is split between a general-purpose processor and one or more specialized, problem-specific processors. Heterogeneous computing is a broader research area and the concept has been around for a while now, it also encompasses efforts like GPGPU computing , the Clearspeed accelerators and more recent efforts like the AGEIA PHYSX co processor .The most recent and the best example of a heterogeneous processor would be the STI Cell B.E with its Power PC core and the 8 synergistic processor units . AMD has plans for a similar architecture and their first heterogeneous multicore offering may be a CPU and GPU in the client space.

On the software side ,task-level parallelism and workload partitioning continue to be the dominant issues for multi-core platforms for both heterogeneous and homogeneous architectures. These issues will be more acute on heterogeneous multi-core systems, since the specialized processors will throw up a new set of problems. I believe heterogeneous computing is geared towards extreme performance computing , the GPGPU movement and the Cell’s performance with folding@home proves this point. General purpose computing might go the homogeneous way , as the challenges are far lesser and are fast being resolved .

I happened to stumble upon this at tgdaily . Its called the Optimus Maximus keyboard , and it is developed by a Russian company called Art Lebedev studio. Now for the cool part of the story , every key of the Optimus Maximus keyboard is a stand-alone display showing the function it is currently associated with .

Other disturbing features of this obscenely expensive piece of hardware are as follows, a standalone LED in each of its 113 keys. Each LED has a size of 10.1 x 10.1 mm and offers a resolution of 48 x 48 pixels. apparently the keys can not only display images, but videos with frame rates of up to 10 fps as well. up to 65,536 colors are supported, which can be seen on viewing angles of up to 160 degrees. Image and video layouts are stored on SD cards, which can be inserted on the back of the keyboard.

Hmmm.. really impressive , if you think about it , its quite an achievement but again , how many computer users ever look down at their damn keyboards while typing . I don’t know many , and for $1500 i ll build a kick ass rig complete with disco lights . Instead of this why not spend all the effort in making keyboards more ergonomic or durable , or try to put in features which make it easier for people with disabilities . This goes up right up there with the Finger Nose Hair Trimmer in my list of utterly useless products . Anyone who can give me one good reason to buy the optimus maximus gets a candy. any takers ??

May 21 saw the annual PCI Special Interest Group’s developer’s conference in San Jose , California . It seems that the move to PCI-E 2.0 is going to happen very soon with a lot of major players showing off PCI-E 2.0 technology at the conference . For the uninitiated PCI-E 2.0 has been in development for some time now and aims to double the interconnect bit rate from 2.5 GT/s to 5 GT/s . It effectively increases the aggregate bandwidth of the 16-lane link to approximately 16 GB/s.
Intel who promised to launch a PCI-E 2.0 motherboard before 2008 rolls in demonstrated unreleased AMD and NVidia graphics chips on its Stoakley chip set for workstations which offers two PCI-E 2.0 ports supporting 16 parallel lanes each. Majors like ARM, LSI, NEC and Synopsys also showed off their PCI-E 2.0 technology at the conference .
Intel is expected to release its first chipsets supporting PCIe 2.0 in the second quarter of 2007 with its ‘Bearlake’ family. AMD will start supporting PCIe 2.0 with its RD700 chipset series and NVIDIA with their MCP72 chipset .The PCI SIG is already working to define a version 3.0 for Express that could appear in products in late 2009. it will probably target 8 or 10 G transfers/second.

So come 2008, get ready to embrace express 2.0 as the new standard , also gear up for faster , high performance graphics cards which eat will eat up 300W of power on the Express 2.0 specification . How fast will the transition happen ? , looking at the merciless move from PCI to AGP and then to PCI express , I would say soon … very soon

I am back , after an amazing trip to Goa and a few trips to the hospital owing to a horrendous bout of viral fever . I really needed this break to get away from my boring daily schedule and get some time for myself . I guess it worked cause my energy levels are on an all time high. So its back to the usual now , at least for the next two months cause then I move on to do my masters.

Recently the One Laptop Per Child (OLPC) group announced that its low-cost laptop would be raised from $100 up to $175 , Closer to home a Chennai based company has introduced a PC for INR 4500 ( Around $100) and is all set to market it to a potential 10 million customers across the Globe , Net PC as they call it is a network computer, designed on a completely new hardware platform without using any of the typical PC or thin client components. The hardware design instead uses components designed and developed for advanced electronic and digital devices. The devices come as a single or a dual processor solution and can be connected to a basic home TV which serves as the display. For as little as $10 a month the company provides access to all the basic computing functions and also provides a broadband connection . There is no local storage though and all storage is done on a remote server maintained by the company . I won’t get into the technical details of the product , visit the Novatium site for more information ,you can download the product specifications there . I believe this is a great initiative especially for developing countries where a PC is still a luxury and not a necessity . Maybe the OLPC guys can take a few hints from this

Computing has come a long way , remember the cray-1, it could do one Gigaflop . Recently AMD announced Teraflop in a box : One Opteron processor and two R600 Gpu’s combining to dish out more than 1 trillion floating-point calculations per second using a general “multiply-add” calculation . Intel at their Beijing IDF showed off their concept 80 core teraflop processor . The Cell B.E is a multicore processor of sorts with one PPE and eight SPE’s capable of crunching out 256Gflops . Teraflop scale processing is already happening on the cell with folding@home and the PS3 and according to reports its showing some great results.

With the industry moving towards this parallel processing revolution , the development on the software side of things seems to be remarkably slow , What I mean here is that the software which uses these multicore processors should be optimized for the same , Most of the ISV’s don’t have in-house programming expertize to do multithreading applications.Switching from one core to many core presents its own set of developmental , and debugging challenges and a large percentage of mission-critical enterprise applications are not “multi-core optimized ” leading to applications not showing any kind of performance boost when switched to multicore processors and in some cases showing even poorer performance, this happens because a single-threaded application can’t utilize the additional cores in the processor efficiently without sacrificing ordered processing. This results in a huge drop in performance due to cores being not optimally utilized . Currently there are no automated compilers for multicore processors , then there is the problem of application priority, there are many such issues which need to be sorted out before the transition to multicore can be complete.

Moving further,The Cell B.E is going mainstream slowly with IBM announcing a new line of servers powered by the Cell B.E . Researchers have already released initial details about the EDGE processor architecture , which stands for Explicit Data Graph Execution.Instead of one instruction at a time, EDGE handles large blocks of data all at once. Using many copies of a small number of replicated tiles, the target for TRIPS ( the first prototype chip on the EDGE architecture ) by 2009 is to hit 5 TFLOPs on 32nm manufacturing. Intel is also looking at Larabee to give them similar numbers.
In the future , there is a good chance that we will be moving to some non X-86 ISA processor , which does the job in a better more efficient way.

More on EDGE and TRIPS
A whitepaper on EDGE

I am a staunch supporter of Sony and I am appreciative of the fact that they have put in a lot of money and time to deliver a machine which is truly outstanding . Every day I read about the Cell B.E breaking new ground , proving again that the PS3 has a lot of horsepower which is currently underutilized . This is an argument I hear from the fanboys a lot , they say that the PS3 packs more power , has the next generation blu-ray drive and has the mother of all CPU’s in the Cell, and in the next two years it will kick everybody’s butt from the XBox360 to the Nintendo Wii and even HD-DVD . I seriously hope so because I wish well for the PS3 , but consider this ..
The PS2 currently has an install base of around 110 million and that is a monstrous number , if the PS3 had been launched alongside the Xbox 360 as a competing console in the same price range rather than a year later , The number of Xbox 360 ’s sold would have been less than a million till date.I am sure 90% of the people who own an Xbox360 now have owned PS2’s before , and they bought their 360’s because there was nothing available in the market which could compete with it, If they had a better option from Sony they would have bought that. So was it a good decision to include the blu-ray player ? it would have been if Sony could deliver it on time and at the same price range of a 360 . but once they realized that it would not match the competition from the cost perspective and also would be late to the market , they should have gone with the standard dual layer DVD format , once they had the blu-ray diode mess sorted out they could have launched another version with the blu-ray for a higher price point . All this time without a competitor has given MS a good install base which third party developers just can’t neglect . If multiple platforms for DMC4 can move another 1 million units for Capcom then it makes sense to them financially, because in the end its all about the money .
Another factor is the cost .. here in India a PS3 costs Rs40000 ($1000) and an Xbox 360 basic costs Rs20000 ($500) , the difference is humongous , The consoles don’t sell in the millions here but if someone wants to buy one which one would it be ? any guesses !! . But I still have hope for the PS3 because all rumors point to a price drop of at least $100 , starting in June, the blue-violet laser diodes will only cost 900 yen to produce (about $8). Last year, the Blu-ray player itself was estimated to cost $125 — a good chunk of that was probably the diodes.
And that brings me to the last and most important factor - content , the DS sells far more than the superior PSP because it has excellent content available for it , the iPod sold 100 million because Apple has developed an entire ecosystem around the product , and likewise PS2 sold millions because of the content which came out for it over the years . When the demand for PS3 content increases , the PS3 will start to sell more.

I wouldn’t count the PS3 out because its an amazing piece of hardware and has had a good reception in most places it has launched , I won’t be surprised if it indeed emerges as the next gen leader

nVidia has strapped on 2 Quadro FX5600 GPU’s ( OpenGl version of the mighty 8800GTX) onto thier already impressive line of Quadro plex VCS ( Visual computing system) . Looks like they are getting ready to take on the emerging GPCPU market as the new workstations support GPU computing with nVidia’s CUDA programming.
GPCPU has become a lucrative market for the future and a lot of big names are pouring money into it , recently Rapidmind , a startup grabbed $10 million of VC money for their GPCPU platform, keeping in pace nVidia is pushing its graphics cards hard as a platform for massively multi-threaded processing applications.
With the new GPU’s, the total frame buffer goes up to 3Gb (1.5 Gb per GPU) FSAA (full screen anti-aliasing) goes up to 64X , and as with the 8800GTX , the GPU’s have a unified shader architecture and fully support shader model 4.0 . Now comes my favorite part - performance stats 64x SLI FSAA,16 synchronized output channels,8 HD SDI channels,60 billion pixels/sec fill rate,1 billion triangles/sec geometry performance and is able to handle up to 148 megapixels display walls. and you can have many of these Quadro Plex boxes in your visualization cluster for scalability. Oh ! I forgot to mention they cost $18000 a piece.