Well, let me just say that I don't see how anybody would even notice a lack of 2D clockspeed in a desktop mainstream part. It would be a killer for a laptop part, but somehow I doubt that nVidia would even try to sell a chip as a laptop part that had a major bug like that in its power savings circuitry.
The Official G84/G86 Rumours & Speculation Thread
- Thread starter Arun
- Start date
We're talking about an excess of 2D clockspeed. And you'd notice because the cooling fan on your graphics card would be spinning at full throttle even when you were just browsing the web. You'd probably also notice when you got your electricity bill.
We're talking about an excess of 2D clockspeed. And you'd notice because the cooling fan on your graphics card would be spinning at full throttle even when you were just browsing the web. You'd probably also notice when you got your electricity bill.
Unless the fan is so quiet it doesn't really need to throttle down that much.
Aerows
Regular
513 is what mine says. Of course, that is what nTune claims, so... It's a Geforce 8800GTS.
Well, for channel I'd agree. OEMs might not be thrilled.
The 8800 GTS is clocked at 500MHz by default, no? The point is, there is no "2D clock" in G80, at least one that I have ever observed. So what the hell is Charlie talking about? Seems like he is covering his ass regarding his report of a "delay" with "no delay, but don't buy them, the are defective!".
G80 has that effectively managed by its clock domains combined with "wide/slow ROPs". Do we know that's the case with the midrange parts?
I think something being overlooked is that G80 does support different performance profiles in the BIOS, whether shipping BIOS revisions have them present or not. Maybe current mobile G8x parts are having BIOS revisions sent out with one profile.
As long as the hardware's not broken in that regard (be it a power issue or something else), I can't see why mobile G8x products can't support it if needed.
As long as the hardware's not broken in that regard (be it a power issue or something else), I can't see why mobile G8x products can't support it if needed.
1. 2D uses less power anyway, even without throttling the clock. I don't see why the fan shouldn't spin down.We're talking about an excess of 2D clockspeed. And you'd notice because the cooling fan on your graphics card would be spinning at full throttle even when you were just browsing the web. You'd probably also notice when you got your electricity bill.
2. Your normal power bill will probably fluctuate more than the added cost of running at a higher clock in 2D. That's not to say you won't pay more, but the cost won't be obvious.
But, in any case, I see no reason whatsoever to expect that this is actually going to occur. Even if it is a real report, it could easily be fixed before launch (it might be something as simple as a driver issue, after all).
Well for what it's worth, I also heard from a reliable source that the midrange cards were having some problems. Many stores already have the cards and are waiting for launch, but nVidia had to rework the cards at last minute because they were getting too hot, which seems to imply that the Inq was right this time.
And God bless him for it
It's a welcome change of pace from the free pass everyone else gives NV.
Fudzilla's take on the situation:
Link: http://www.fudzilla.com/index.php?option=com_content&task=view&id=385&Itemid=1
Link: http://www.fudzilla.com/index.php?option=com_content&task=view&id=385&Itemid=1
vertex_shader
Banned
Same thing happend in the 8800gts/gtx launch, many of this "refurbished" cards dead because the distributors service changed the affected part, and not NV.
First things first, looks like I own Charlie an apology.
Huh? As I recall, this was only an issue for people who got their cards prior to official release.
Huh? As I recall, this was only an issue for people who got their cards prior to official release.
Aerows
Regular
A few people received cards without the change resistor, transistor, whatever the part was that needed the change. I saw posts with pictures. Whether they got them prior to release or not, I can't answer that question, but some people did get them.
Correct me if I am wrong, but I thought that it was only the GTX model that had the issue.
Oops.
Seems like someone decided to do a little early comparison between a 7600 GT and a 8600 GT (not the faster 8600 GTS version, mind you):
http://forums.vr-zone.com/showthread.php?t=141089
Note: This G73 (7600 GT) appears to be the B1 revision, made at 80nm, not the original 90nm one.
Seems like someone decided to do a little early comparison between a 7600 GT and a 8600 GT (not the faster 8600 GTS version, mind you):
http://forums.vr-zone.com/showthread.php?t=141089
Note: This G73 (7600 GT) appears to be the B1 revision, made at 80nm, not the original 90nm one.
Similar threads
