Is to be tested. Since the chips were launched yesterday (Sept 24, 2003), it would be hard to have a thourough test. I will admit that Opteron has been out for some time, but doubt that enough Mandrake users have one to test it enough.
This was true for the past umpteen years, but lately ATI and Nvidia have been designing seperate chips for mid and low end sales (the most famous being the 9500, a crippled high-end chip vs. the 9600 which was designed to be smaller and wimpier than its big brother).
The real catch is that the mid-range market is built in to the motherboard. Presumably Via bought S3 for this, but S3 seems to want to aim for the sky (sort of).
The thing that hasn't changed is that all big graphics companies got that way by challenging at the high end. S3 was there when they basically brought acceleration to PCs (for you whippersnappers, this was in the win 3.0 era). 3dfx reigned king, but couldn't replace the voodoo architecture, NVidia wrested control via TNT2, ATI was steadily losing market share until Radeon, I can't see how a "mid range" strategy is going to live. S3 is basically betting the company that ATI and NVidia will collude to attempt to keep prices artificially high, and then sweep away the midrange. Can't guarentee it will happen (just ask cyrix and winchip (also owned by Via) about trying to eat the crumbs of giants).
Nothing taught the horror of spegetti code like trying to cram 48k of basic code into 32k (after "dos") RAM. On an Atari 800 (basic) you had 128 variables, run out of them at your peril. Contants ate up more memory than variables. Of course, your initializtion routines could be overwritten with a little work (of course, you have something like: "press enter to continue/anything else will end the program".
On the other hand, visual basic 2.0 would probably fit the bill. It worked fine, and didn't try to dump the whole visual studio on you at once.
If you read the industry mags, you will note that all the analysts basically say "get your act together and start fixing prices like everyone else". The DRAM industry is probably the only business in the US that actually allows a free market, and this is considered completely unnatural.
Sooner or later they will manage to fix prices, and you will be able to tell by consistant profits by memory manufacturers.
Remember the first lesson in business 101 is never be forced to compete. Read Warren Buffet's advice for stock picking: you want a business with a "franchise" that allows it to prevent competitors from eating their lunch, thats where the profits are.
FPGAs are great for complex control logic in hardware. They can also be used for DSP functions.
Using FPGAs for general purpose computing efficiently is difficult. You essentially start out with a 100:1 handicap against a commodity CPU (this comes from the amount of transitors per gate, using VHDL vs. custom design, and the way routing wires have to be universal).
Programming such a beast becomes an exersize in finding huge amounts of parallelism that don't require memory accesses (FPGAs have limited RAM on board, but don't expect much, and you have to share accross 100s of functions).
Supposedly there are C to hardware complilers out there, but I can't see Joe Software designer chugging out code that carefully checks to see how every line affects the clock rate (remember that: in software you have 10% of the code executed 90% of the time, in harware you have 100% of the code executed 100% of the time. The FPGA can only clock as fast as the slowest path).
The economics are probably the worst problem. These sort of things are most likely to go into government or military instalations where the contract says hardware has to do (impossible thing), and be maintained for x years. The device gets made, the customer changes the requirements, spin repeat until it ships as an expensive system that a simple desktop could do by shipdate. If you build a beowulf to do something, with a little forsight you can turn around in two years and double the power. With an FPGA design, you may be able to buy an off-the-shelf board that has advanced (if the company is still in buisness, not guarenteed), but then you have to dig into the source and modify for the new chips. This gets expensive fast.
Slashdot Mirror
Wumpus
The real catch is that the mid-range market is built in to the motherboard. Presumably Via bought S3 for this, but S3 seems to want to aim for the sky (sort of).
The thing that hasn't changed is that all big graphics companies got that way by challenging at the high end. S3 was there when they basically brought acceleration to PCs (for you whippersnappers, this was in the win 3.0 era). 3dfx reigned king, but couldn't replace the voodoo architecture, NVidia wrested control via TNT2, ATI was steadily losing market share until Radeon, I can't see how a "mid range" strategy is going to live. S3 is basically betting the company that ATI and NVidia will collude to attempt to keep prices artificially high, and then sweep away the midrange. Can't guarentee it will happen (just ask cyrix and winchip (also owned by Via) about trying to eat the crumbs of giants).
Wumpus out
Steadily decreasing box size, I smell a problem.
Wumpus
no text
If you are under two years of age anyway... Thats where the legend came from. Wumpus
On the other hand, visual basic 2.0 would probably fit the bill. It worked fine, and didn't try to dump the whole visual studio on you at once.
Wumpus
Sooner or later they will manage to fix prices, and you will be able to tell by consistant profits by memory manufacturers.
Remember the first lesson in business 101 is never be forced to compete. Read Warren Buffet's advice for stock picking: you want a business with a "franchise" that allows it to prevent competitors from eating their lunch, thats where the profits are.
Wumpus
It might have been, but the afterburners never turned off.
Nine rings for (faulty) ram chips doomed to die. Wumpus
FPGAs are great for complex control logic in hardware. They can also be used for DSP functions. Using FPGAs for general purpose computing efficiently is difficult. You essentially start out with a 100:1 handicap against a commodity CPU (this comes from the amount of transitors per gate, using VHDL vs. custom design, and the way routing wires have to be universal). Programming such a beast becomes an exersize in finding huge amounts of parallelism that don't require memory accesses (FPGAs have limited RAM on board, but don't expect much, and you have to share accross 100s of functions). Supposedly there are C to hardware complilers out there, but I can't see Joe Software designer chugging out code that carefully checks to see how every line affects the clock rate (remember that: in software you have 10% of the code executed 90% of the time, in harware you have 100% of the code executed 100% of the time. The FPGA can only clock as fast as the slowest path). The economics are probably the worst problem. These sort of things are most likely to go into government or military instalations where the contract says hardware has to do (impossible thing), and be maintained for x years. The device gets made, the customer changes the requirements, spin repeat until it ships as an expensive system that a simple desktop could do by shipdate. If you build a beowulf to do something, with a little forsight you can turn around in two years and double the power. With an FPGA design, you may be able to buy an off-the-shelf board that has advanced (if the company is still in buisness, not guarenteed), but then you have to dig into the source and modify for the new chips. This gets expensive fast.