DeltaChrome details

demalion

Veteran
Tech -Report posted some interesting specifics on the DeltaChrome, and the part is looking like a fair contender.

Particulary interesting to me was discussion about DX 9 capabilities, including specific diagrams about what "pipelines" mean for the card, and information on what functionality is exposed.

With a timely release, I think we have a feasible 3 contenders at the least. The sacrifices seem feasible and wise tradeoffs for their target market, and the amount of forthright info is remarkable.

Certain enigmatic persons could benefit from imitation. Kriscoughtof andahemfriends

Maybe XGI will be inspired as well.
 
Looks like ati won't need to worry about losing the gamer market to DeltaChrome. These won't be hugely popular for hardcore gamers but there's a big market for affordable cards for casual or non-gamers.
 
What a throwback, here was me thinking the market finally realized edge-AA was simply too cheap in transistor count to ignore and there comes S3. Hopefully IMG will not similarely disappoint me ... oh well, it is their funeral.
 
I's not fair I post the news and hour earlier and I'm the one who gets locked, poo, poo, poo. ;)

Well It's looks like quite a reasonable part to me. Slightly disappointed by the memory interface (no crossbar) and AA (only 2!) but could still find a way into a 2nd system of mine.
 
I pretty much agree with Heathen. Both this and the XGI parts look suprisingly competent, with no real huge flaws or obvious problems (aside from the 2x AA maximum in this part...).

Performance wise they could certainly get a market in the mainstream segment, but the lack of brand name recognition could be a problem. Still, this is a solid start and hopefully will provide competition to the big 2 for some time to come :)
 
but the lack of brand name recognition could be a problem

That should hopefully drive the price down so it falls into my 'impulse buy' range.

As long as S3 can get reasonably solid drivers on the first try around it's got a definite chance for the casual gamer and some OEM wins.
 
Heathen said:
I's not fair I post the news and hour earlier and I'm the one who gets locked, poo, poo, poo. ;)

Heh, some of my bad thread-starting karma must be balancing out. 8) Your turn to be jinxed.
 
how can it be limited to 2x super-sampling? Can't you just up the render res and then downsample? (probably not very useful for performance reasons, and would really benefit from per-pixel jitter, but...)

Serge
 
psurge said:
how can it be limited to 2x super-sampling? Can't you just up the render res and then downsample? (probably not very useful for performance reasons, and would really benefit from per-pixel jitter, but...)
Mightn't there be limitations on how large a surface you can render to is?
 
Heathen said:
but the lack of brand name recognition could be a problem

That should hopefully drive the price down so it falls into my 'impulse buy' range.

As long as S3 can get reasonably solid drivers on the first try around it's got a definite chance for the casual gamer and some OEM wins.

Yeah a third party in there will really add to the competitive nature of the market possibly forcing ATi and Nvidia to become even more competitive in their high and low end parts. In reality though S3 will have a time getting any decent margins on the Delta-Chrome core though. It is as Heathen suggested they have absolutely no brand name recognition. This was a tuff lesson for ATi to learn. OEMs are most interested in brand name identification to help sell their boxes. These Delta-Chrome chips would really have to lay a hammering on both ATi and Nvidia just to start towards making a name for themselves. Of course I am presumptuous here but I do not really think that they will have that sort of impact but who knows. All we have seen of these so far is basically a 10 month old announcement with a few details no performance numbers or anything. Optimistically, they do make some inroads with the chip but I suspect their supposed midrange card will come out smelling like a lower end solution and priced likewise. There will not be any in-roads on the high end at all and likely their higher end solution will be priced mid range and perform on par with ATi and Nvidia mid-range solutions. I would be stunned if indeed they do end up trouncing any of the current high end solutions from ATi and Nvidia. S3 though may make some real inroads on the low end market nibbling on Nvidias OEM 5200 deals.
 
It looks good. If the price is right, this could make for a very viable card.
 
Pete said:
It looks good. If the price is right, this could make for a very viable card.

Sure it may to some enthusiast if it cuts the mustard but as far as being an out and out success with OEMs it is nearly a sure bet that they will be religated to lower end models. S3 simply does not have the brand name recognition it takes to get decent margins or for that matter market share, yet. Sure if they release another good chipset next year or in six months (assuming that this one is "good enough") they might start to make a dent in market share etc. I am however very skeptical of this chip, all we can do is wait and see.
 
About FSAA

The DeltaChrome presentation that was put on their website earlier this year and the current one says:

“Arbitrary-x super sampling and 16x multi-sampling high performance anti-aliasingâ€￾

I've e-mailed TR and asked them if they could double check their info.
Update: Scott, the uthor of the article, replied and said that they made it more than clear to him that it would only support SSAA and only 2x.
Awhh crap, there goes that chip down the drain. I mean the model supposed to take on 5900 non-Ultra just got utterly uninteresting../Update

If I remember corrently their first PDF file even had before/after picture showing their FSAA method. I deleted them from my hard drive yesterdayt though since I thought they wouldn't be of any use since they had the new files up there. :rolleyes:
 
I may be stupid as I don't understand the excitement surrounding the Deltachrome as we already have a third contender (XGI) with a series of cards (Volari) which look far more exciting than the Deltachrome as Volari doesn't have this 2x AA maximum limitation

Don't get me wrong, I'm happy of course to see new contenders in the 3D arena and always have been in favor of seeing competition at work, but the Deltachrome just doesn't excite me that much, especially since they don't have an A1 silicon done yet.
 
xGL said:
I may be stupid as I don't understand the excitement surrounding the Deltachrome as we already have a third contender (XGI) with a series of cards (Volari) which look far more exciting than the Deltachrome as Volari doesn't have this 2x AA maximum limitation

Don't get me wrong, I'm happy of course to see new contenders in the 3D arena and always have been in favor of seeing competition at work, but the Deltachrome just doesn't excite me that much, especially since they don't have an A1 silicon done yet.

comparing deltachrome to volari:

Both somewhat lack bandwith ( 128-bit bus)

volari lacks shader power ( only half of pipes have shader )

deltachrome is coming earlier than volari

deltachrome has better video options

volari has 2-chip option.

then this deltrachrome AA confusion.

If deltachrome did MSAA it would be much moreo interesting chip than volari for me.
 
Back
Top