• We’re currently investigating an issue related to the forum theme and styling that is impacting page layout and visual formatting. The problem has been identified, and we are actively working on a resolution. There is no impact to user data or functionality, this is strictly a front-end display issue. We’ll post an update once the fix has been deployed. Thanks for your patience while we get this sorted.

Will we still see Rampage with 3dfx being bought?

I very much doubt it. The Rampage and all other future 3dfx cards are effectively dead. At best we may see some elements of the technology planned for those boards inplemented in future NVidia boards.
 
I want to know how close Rampage was to being finished. It sounded like it was due to be released in the very near future. If the product is ready to roll, then there is a good chance nVidia would release it, I would imagine.

Then again, if it were ready to roll, it would be very odd for 3dfx to fold at this time.
 
Don't forget how the V5 was ready last year, however production problems and bugs hit it, so it was too late for its own good...
 
AMD bought NextGen, and made their Nx6x86 CPU the AMD K6. Perhaps we'll see a nVidia Riva MX or something like that. According to rumours chip's quite far in developement; it'd be foolish for nVidia to throw it away. Especially since being multichip capable it has the ability to cover also lower market segments which high-end expensive NV20 chip can't.
 
jprod - I both agree, and disagree with you 🙂

What I agree with:

That Rampage was far enough along in development for, at least at first glance not to be worth canceling. I've heard it said that is why 3dfx didn't include the Gigapixel technology into it, because they had already invested so much, and while it would be better, it would also delay the product to redesign it at such a fundamental level.

Multichip could be an interesting proposition for NVIDIA...though they have publicly bashed the > 1 chip solution.

What I disagree with:

I don't think they'll use Rampage directly (perhaps some of its innovations, whatever they are). Your comparison between the AMD/Nexgen situation is a far cry from the current situation. AMD was wicked late in developing their K5, and because of that, the performance wasn't on par (because it couldn't ramp well enough - it was great per clock, but horrid in ramping). AMD needed Nexgen to even continue as a company. They are lucky that they had the cash to do so, or else we'd have no more AMD in its current state.

This is different: Nvidia doesn't need the technology that 3dfx had per se, except perhaps for the technology that Gigapixel had. I mean, sure it helps, both in the fact that it eliminates a competitor and strengthens them technilogically, as well as in the patent deptartment, but its not like Nvidia is "on their last legs" so-to-speak like AMD was with the K6. The K6 allowed AMD to have time to develop the K7 (and the talent to do so with Nexgen engineers, plus Dirk Meyer from API). Without it, they'd be dead.

 
That's too bad. The Rampage has been taken too long to develope, it and its maker is going down the toilet. So much for Penstar prediction of a 3dfx come back.
 
I`m sure there are going to big discussion meetings at Nvidia on future products & new technology they get from 3DFX ,I`m now confused as to what next Nvidia card to buy.I think they will take the best ideas from the Rampage & make there own version.

I wish I could be a fly on the wall at those meetings.

🙂
 
BurntKooshie: True, AMD's situation was quite different back then compared with nVidia's current one. But I was more trying to point out that buying out a company by it's competitor doesn't necessarily mean the demise of their upcoming products.

nVidia doesn't need 3dfx tech to compete in near future in high-end market segment, but in a single-chip Rampage design nVidia would have an extremely powerful value solution for near future. NV20 chip might be too large and compex to dribble down to value segment; Rampage is - supposedly - smaller, cheaper to manufacture (I recall 25 million transistors complexity, targeted .18m process) and implements full DX8 feature set which is a great marketing asset, especially with OEMs.

All really depends on how ready Rampage is to go, and on that point we can only speculate 🙂
 
You could be right on this one, jp. A single Rampage chip videocard could aim at the budget-minded consumer, while the NV-20 is aimed at hardcore gamers, until the cost of the NV-20 can be brought down. But I fear that 3dfx has cancelled the development of the Rampage, it probably be too expensive for nvidia to crank it back again. I think the Sage chipset is dead also. In a war between SLI and single-chip solution, the single-chip solution has won, yet again. It took too long for the VS-100 chip to reach the market, which brought 3dfx to this sad ending.
 
No way they will release it....cause it's probably better than NV20, and it would make them look bad because if 3dfx had lived to release it it would've crushed them :|
 
Maybe Nvidias next card will be the rampage, after the NV20.
Combine 2 video cards to make 1 sweet one.
It can only get better now.
Maybe they will let each other go their own way and have the resources to develop it.
This would only increase revenue for Nvidia if they both develop chipsets and let the best one win.
I doubt Nvida would just close the doors on 3Dfx
They want something that 3DFx has got and it must be good.
 
Nvidia probably want the Mosaic technology. 3dfx has not talked much about the Rampage or Sage, and had probably killed it a long time ago. 3dfx was hoping for the Mosaic to rescue the company, but it was too late.
 
Im pretty sure Nvidia will release rampage or nv20(whichever one performs worst) and use the best for "next generation".
 
I seriously doubt that nVidia would use the Rampage. Why? Driver support. I'd seen some flowchart somewhere that showed how nVidia designs into their chips a driver core of some sort that makes writing the initial drivers for their new chips much easier and quicker. That's why when they release a new chip they can get drivers out very quickly and optimize them later. Unlike ATI, for example, where new chips = completely new driver sets = delays. It's one of the smart things nVidia does.

So, if they use Rampage, what are they going to do about drivers? It looks like they haven't hired people from 3dfx, and now they'd have to have break off or hire another group to develop drivers for Rampage. I'd expect it would be cheaper to produce the MX/value version of the NV20 than go through that mess. And which mfgs today have multiple products that use substantially different drivers? 3dfx didn't, since the V4500, V5500, and for that matter the V3 could all use the same driver set. Neither does ATI, since the Radeon is the new basis for its mobile, value and premium lines, and the Rage series is pretty much dead. Matrox uses unified drivers for its G-series. And obviously nVidia doesn't since TNT's and all Geforces use a single driver version.

nVidia bought themselves out of a lawsuit real cheap, and as a bonus killed off a competitor and got some interesting technology. As an aside, Brian Burke stated that 3dfx is still responsible for existing driver support. Which is interesting, because 3dfx is getting rid of basically all of its employees. It's really NOT good news for Voodoo owners today.
 
Chances are, if this sale is allowed to happen, it'll kill the graphics market as we know it. I highly doubt nVidia will use any of the 3dfx technology, except possibly the real high end stuff, which is way beyond what nVidia has (32-way VSA-100 SLI solutions, licensed to Quantum 3D). But this will definately allow nVidia to charge higher prices, and will most likely slow down development of faster cards. I wouldn't be surprised if after this nVidia refreshed the GF2 to release it as the GF3 and saved the NV20 core for the next product cycle. The only things I'm hoping for is that either the govenment doesn't allow the sale, the shareholders don't allow the sale, or Matrox steps up production of the G800.
 
why not deffered rendering from 3dfx from gigapixel? What's the alternative? A brute force, more expensive method of doing 256bit wide DDR (twice as many traces, meaning more PCB layers, etc.)

Or, they could do embedded dram. Die size then becomes an issue, and costs rise for that reason, plus the fact that they have litte/no experience with embedded dram.
 
Ummm hello? Are you all idiots? Companies do NOT buy other companies and not use their patents/technology/R&D. I mean Nvidia is buying the core of 3dfx(their chip technologys etc). 3dfx has spent tons of money on R&D Nvidia won't just flush it all away. You all forget, 3dfx owns Gigapixel. There are a number of 3dfx technologies Nvidia will use, they would be stupid not to.
 
Ummm hello? Are you all idiots? Companies do NOT buy other companies and not use their patents/technology/R&D

1) Ummm hello? We're talking about the Rampage chip (see thread title). Rampage is an actual chip, not a patent or non-applied technology. Read my post above about why it may actually be cheaper to throw away Rampage.

2) Flat out untrue. Sometimes it's a lot easier to buy a company than go through protracted litigation. Further, wasn't the 3dfx suit over multitexturing? Assuming 3dfx won, how much do you suppose 3dfx could have extracted from nVidia over a lifetime if nVidia had to pay royalties on each chip they've ever made since the patent that used multitexturing, including back royalties on each Geforce/Geforce2/Geforce2 MX/Geforce2Go chip they've ever made? Also, if nVidia gets those patents wouldn't they make a convenient weapon against, oh say, ATI? nVidia doesn't have to implement anything from 3dfx to make this a good deal. If nVidia had wanted only Gigapixel technology they could have had it for a heck of a lot less than ~$110 million. I believe 3dfx only paid $16 million for Gigapixel.

3) nVidia's actions in regard to 3dfx departs from its usual course of business - hiring away key engineers. They did it to Aureal just before Creative bought Aureal's remains, and they've allegedly raided Matrox's engineers, which is the rumored reason for the G800's delay.

BTW, there's another case of buying off lawsuits. Creative sucked Aureal dry with litigation costs before Aureal won, so what was $28 million or so for Creative to buy up Aureal and eliminate both the problem and the competition?
 
You must log in or register to reply here.

TRENDING THREADS

Back
Top