backtop


Print 50 comment(s) - last by FaceMaster.. on Mar 17 at 6:53 PM

Marchitecture at its finest

NVIDIA has the fastest single GPU for the desktop in the GTX 285, a 55nm die-shrunk version of its predecessor the GTX 280. However, ATI has been able to gain a larger market share due to aggressive pricing and ramping of smaller geometries.  This has led to price pressure on NVIDIA, especially in the performance mainstream segment.

NVIDIA's original GT200 chip -- which is used in the GTX 280 and GTX 260 -- is too big, too costly, and consumes too much power to be used effectively in a mobile solution. NVIDIA has already switched to TSMC's 55nm process from the baseline 65nm node to deal with these issues for the GTX 285, but it is still not suitable for the majority of laptop users. Battery life is too short, the cooling fan is too loud, and the cost is too much.

One solution was to begin manufacturing on the 40nm bulk process like ATI has done. According to our sources, NVIDIA's attempts to produce a die-shrunk 40nm GT200 chip were "disastrous at best". Design problems became evident, since the GT200 was originally designed for the 65nm node. Two shrinks in a row without a major redesign was just too much for NVIDIA, and our most recent information from Taiwan is that the first 40nm chips from NVIDIA will be in the GeForce 300 series.

Without a power efficient GT200 based GPU solution for the mobile or mainstream value markets, NVIDIA is rebranding the 55nm G92b chip yet again to meet these critical segments. The original 65nm G92 chip was used in the GeForce 8800 GT, but you can only do so much with an older design. The chip was respun as the G92b with a 55nm die shrink, and is currently used in the 9800 GTX+. All G92 chips are only DirectX 10 capable, and will not support the full feature set of DirectX 10.1 or DirectX 11 that will come with Windows 7.

The problem is that many consumers will pick up a GTX 280M or GTX 260M thinking that it is the same or similar to the GTX 280, when it is actually just a 9800 GTX+.

There is currently no GeForce 100 series for the desktop market, but Nvidia launched the GeForce GTS 160M and 150M for notebooks at the same time as the GTX280M and the GTX260M.



Comments     Threshold


This article is over a month old, voting and posting comments is disabled

RE: Ignorance is bliss
By boogle on 3/5/2009 5:51:49 AM , Rating: 1
If I release product Z that's faster than product A - does it matter which one has what underneath?

In your analogy, if the Phenom Mobile 9550 was faster than the competition, and/or other CPUs from the same brand with the Athlon X2 name - would you still have a problem with the name? What about the Sempron which was just a slightly cut down Athlon XP? What about the Celeron? What if I had a Celeron that was faster than a Pentium? Should I continue to call the Celeron a Celeron? Would it not be more confusing having a 'budget' name beating a proper name? Surely it would be less confusing to just brand the oddly fast Celeron a 'Pentium' purely to avoid performance confusion.

I often think NV should just rename the 'codenames' for these cores so that even tho its still G92 it's renamed to something like GT200M and launched as a 'new' core. Then no one would complain. It's only this teeny bit of additional knowledge that has people going mental. Never mind that G92 is still competitive...


"We are going to continue to work with them to make sure they understand the reality of the Internet.  A lot of these people don't have Ph.Ds, and they don't have a degree in computer science." -- RIM co-CEO Michael Lazaridis














botimage
Copyright 2014 DailyTech LLC. - RSS Feed | Advertise | About Us | Ethics | FAQ | Terms, Conditions & Privacy Information | Kristopher Kubicki