Ascended Saiyan
Newcomer
NV-30 problems ?!!
- Thread starter Ascended Saiyan
- Start date
I don't think the term 'high-power chip' relates to power problems delaying it. It seems they just talking about issues with the .13umn process.
LittlePenny
Regular
It is nice to know the most over-dramatic people in existence handle all the money. Although I am just a bit curious as to how much this chip has been delayed since they were very adament about it being the first DX9 chip.
I will leave that one to pascal to make the obvious porn references.
This article is way over dramatized. I mean, would NVidia really cut research and development into new chips as soon as their profits decline (they have been doubling year over year for the last few years). It's utterly ridiculous. The company is awash in cash and has good credit. The management have been in this business long enough to know that you can't regain your profit position by cutting R&D and not developing anything new. That's a recipe for NVidia going out of business.
The converse is true. A recent item noted their commitment to expanding R&D in the face of competition/contraction in the market.
Has nothing to do with the thrust of their agrument, in that the Intel integrated chipsets will be larger volume sellers. Of course this is a segmented market of Intel's making & as such the two chipsets are currently not perfect substitutes. I make no claims as to profit wedges.
DaveBaumann said:
I think so too... Wasn't it back in November when the first reports about nVidia having issues with TSMC's .13 process came out? I remember that TSMC in April 2002 defended their .13 process and stated that the issues were being adressed and fixed.
Maybe that is what Business News is refering to - or could there really still be problems at TSMC?
If you read any investment board then the 'intel factor' has been talked of for a long time. The issue stems from the fact that until i845G there were no integrated chipsets for Pentium4, and so the discrete graphics vendors have been reaping the benefit of this - the fear was that now i845G was around much of the low end business that they've enjoyed since P4's introduction will be undercut by Intel.
The western allusions were a bit much, but otherwise the article seemed relatively fair, if brief and not too accurate. I think high-power means top-of-the-line (obviously it will consume more power, but they meant it in terms of speed, not power draw). Intel is obviously a competitor in the motherboard segment, but for different CPUs--not the best reporting there. They probably thought Intel was a flashier name than VIA or Sis.
Business Week is a pretty big name in business news. I don't follow the market that much, but I think they're the biggest weekly business magazine, if not the only one. That explains the article date, too--it's for the upcoming July 29th issue.
nVidia might have problems with "full production volumes," but I'm sure they'd prefer to release in limited numbers sooner, rather than in full force later. They're big only because they're known to have the fastest chip, and they need to regain that title quickly.
Business Week is a pretty big name in business news. I don't follow the market that much, but I think they're the biggest weekly business magazine, if not the only one. That explains the article date, too--it's for the upcoming July 29th issue.
nVidia might have problems with "full production volumes," but I'm sure they'd prefer to release in limited numbers sooner, rather than in full force later. They're big only because they're known to have the fastest chip, and they need to regain that title quickly.
G
Guest
Guest
There are good reasons to believe that the nv30 will not see the light of day for sometime. The first and most troublesome issue is obviously the Radeon 9700. Granted the card is running at the .15micron process but ATI has managed through either good engineering or some sort of technological implement to get higher clock rates without heat killing the chip. Nvidia seems to need the .13micron process for the nv30 and as a result is waiting for it.
Further there is some speculation from more then one source that the nv30 will be late. ATi for example expects that the NV30 was initially meant to be nvidias spring part. AnandTech made note of this while reviewing the Radeon 9700.
http://www.anandtech.com/video/showdoc.html?i=1656&p=1
"The folks over at ATI are banking on NV30 being a 2003 part, so that the Radeon 9700 can close out this year with a bang. If you want the absolute fastest graphics card available, the Radeon 9700 fits the bill better than any GeForce4 or Parhelia; and it will continue to do so for months to come."
Granted this is coming from ATI but....... this may very well be the case. Considering that the Radeon 9700 is such a leap in performance compared to anything else that has been put out for some time ATi may very well have cought nvidia with their guard down. The NV28 may have been their next production in line but in light of the extreme power of the Radeon 9700 the nv28 may have gotten thrown to the back burner so to speak in favor of the nv30. Here is where we start seeing problems nvidia knowing full well they are behind ATI at this point announces in their last CC that they have had a "record number of tape outs". While this is normally regarded as a negative one could infer that they certainly are making efforts to do *something* rather quickly at that.
According to papers nvidia is circulating with regards to the nv30 its specs are impressive and from what some have seen in them the nv30 will be faster. But for the sake of argument sometimes things don't always work out in real life as well as they do on paper. But do these papers really say that the nv30 will be available in a few months? I beg to differ. Normally a chip that has just taped out (with the rumored 120 million transistors mind you.) doesn't just magically appear on the market a few months later driver/hardware trouble free. Making a chip of that magnitude with new technology done at the troublesome .13micron process will take more then a few months. In the past nvidia has had quick answers to ATI products. While appearing to have quick execution and product delivery nvidia really has only since 1999 had to big chips created.(The Geforce 1 and the Geforce 3.) The rest of the products they have created were just augmented forms of these two cores. Now they have to develop a new core not just another augmented one. Nvidia managed to have the geforce three appear seamlessly with their product role out as if they had developed it since the Geforce 2 Ultra, but in reality they had been working on that core since at least the announcement of the xbox in fact the geforce 3 is likely a spin off product of the xbox.(Getting OT here.)
To say that the nv30 will be available soon is folly. There is more then one market analyst concerned about nvidias prospects. Some stake their companies reputation on these "opinions" that are often based on knowledge that isn't public info at all.
http://biz.yahoo.com/rc/020718/tech_nvidia_research1_1.html
Nvidias CEO has also stated unofficially (finding were this statement is seems to be quite difficult, and Kyle from HardOCP has stated that nvidia does not officially recognize this statement.) that the nv30 will be available as a fall part. But as we know nvidia said that the nforce chip set would be available last summer but never was made available to the market until november?(maybe it was december not entirely sure but the point is nvidia has not always executed perfectly.)
Who knows maybe the Radeon 9700 is more powerful then the nv30 was originally and as a result nvidia has had to reengineer the part?
Untill we actually see the part in action via previews all this talk of the nv30 being available in the fall is just retoric. Nvidia PR is prolly working overtime to dampen ATIs launch of the Radeon 9700 as much as possible. The nv30 is obviously vaporware but now it seems that the Radeon 9700 is a reality the nv30(vaporware) is receiving sooo much hype that you would think it was just around the corner, even though there really is little evidence to warrant this conclusion.
Further there is some speculation from more then one source that the nv30 will be late. ATi for example expects that the NV30 was initially meant to be nvidias spring part. AnandTech made note of this while reviewing the Radeon 9700.
http://www.anandtech.com/video/showdoc.html?i=1656&p=1
"The folks over at ATI are banking on NV30 being a 2003 part, so that the Radeon 9700 can close out this year with a bang. If you want the absolute fastest graphics card available, the Radeon 9700 fits the bill better than any GeForce4 or Parhelia; and it will continue to do so for months to come."
Granted this is coming from ATI but....... this may very well be the case. Considering that the Radeon 9700 is such a leap in performance compared to anything else that has been put out for some time ATi may very well have cought nvidia with their guard down. The NV28 may have been their next production in line but in light of the extreme power of the Radeon 9700 the nv28 may have gotten thrown to the back burner so to speak in favor of the nv30. Here is where we start seeing problems nvidia knowing full well they are behind ATI at this point announces in their last CC that they have had a "record number of tape outs". While this is normally regarded as a negative one could infer that they certainly are making efforts to do *something* rather quickly at that.
According to papers nvidia is circulating with regards to the nv30 its specs are impressive and from what some have seen in them the nv30 will be faster. But for the sake of argument sometimes things don't always work out in real life as well as they do on paper. But do these papers really say that the nv30 will be available in a few months? I beg to differ. Normally a chip that has just taped out (with the rumored 120 million transistors mind you.) doesn't just magically appear on the market a few months later driver/hardware trouble free. Making a chip of that magnitude with new technology done at the troublesome .13micron process will take more then a few months. In the past nvidia has had quick answers to ATI products. While appearing to have quick execution and product delivery nvidia really has only since 1999 had to big chips created.(The Geforce 1 and the Geforce 3.) The rest of the products they have created were just augmented forms of these two cores. Now they have to develop a new core not just another augmented one. Nvidia managed to have the geforce three appear seamlessly with their product role out as if they had developed it since the Geforce 2 Ultra, but in reality they had been working on that core since at least the announcement of the xbox in fact the geforce 3 is likely a spin off product of the xbox.(Getting OT here.)
To say that the nv30 will be available soon is folly. There is more then one market analyst concerned about nvidias prospects. Some stake their companies reputation on these "opinions" that are often based on knowledge that isn't public info at all.
http://biz.yahoo.com/rc/020718/tech_nvidia_research1_1.html
Nvidias CEO has also stated unofficially (finding were this statement is seems to be quite difficult, and Kyle from HardOCP has stated that nvidia does not officially recognize this statement.) that the nv30 will be available as a fall part. But as we know nvidia said that the nforce chip set would be available last summer but never was made available to the market until november?(maybe it was december not entirely sure but the point is nvidia has not always executed perfectly.)
Who knows maybe the Radeon 9700 is more powerful then the nv30 was originally and as a result nvidia has had to reengineer the part?
Untill we actually see the part in action via previews all this talk of the nv30 being available in the fall is just retoric. Nvidia PR is prolly working overtime to dampen ATIs launch of the Radeon 9700 as much as possible. The nv30 is obviously vaporware but now it seems that the Radeon 9700 is a reality the nv30(vaporware) is receiving sooo much hype that you would think it was just around the corner, even though there really is little evidence to warrant this conclusion.
If they did have to do a new fab spin it could be anywhere from 6 to 12 weeks before they see that part out of the fab. When your talking about IC's manufacturing, "quick" is a realtive term. I dont think thats the case. I think they will stick with thier design prior to the R300 launch.
G
Guest
Guest
jb said:
Yeah I don't disagree with you there at all. That was really a small argument. If they do stick with their original design though it may be that the part does not out shine the Radeon 9700 entirely resulting in disapointment to the market. Further ATI is likely working on the .13 micron process as well and since they already have a successful version working at the .15micron process their transistion may hold fewer problems in the fab at .13microns. I am no chip engineer but that seems like a logical conclusion.
Geek_2002 said:
Armchair analysts should at least get their facts straight before spending all that time writing those words who's end result is drivel.
If you had listened to the conference call (which was publicly available on nvidia.com, but apparently is not anymore), the CEO/CFO was explaining about the balance sheets and in particular about why the engineering costs for the quarter were so high. He stated a record number of different parts that had taped out that quarter, not that one part had taped out repeatedly. No CEO in his right mind would state "oh yeah, we had to tape out our flagship product a record number of times" at an investor conference call.
We obviously know that NForce2 (You could arguably call that 4 products, with the 2 different north bridges and 2 different south bridges) had taped out, we could guess that the other product(s) would be NV30 and/or NV28. Is that enough to be a "record number"? Maybe NVIDIA has some other products about to come out that we just don't know about.
Only (and I'm sorry for being so rude here) dolts who don't understand english well enough to listen and read keep regurgitating this fallacy of "NV30 taped out a record number of times".
G
Guest
Guest
RussSchultz said:
And you are saying that nvidia is having difficulties producing nforce chipset?(done at .15micron process, that is using gf mx technology get with it ??!) Further you are suggesting that nvidia would not have difficulties with the nv30 120 million transistors at .13 micron process? Only a "dolt" as you put it would assume that the process is problem free.
You continue to demonstrate your inability to comprehend the written word.
Where exactly do I state than nvidia is having difficulties producing nforce chipset?
Where exactly do I state than nvidia is having difficulties producing nforce chipset?