The LAST R600 Rumours & Speculation Thread

[Russell]

L'Inq has chimed in, and it's not Fudo!!

http://www.theinquirer.net/default.aspx?article=37880

* Natoma splashes holy water on himself for his L'Inq-referencing transgression

Aggghhhhhhh it burns!!!!!

Charlie tends to be one of the more reliable Inq folks. Though it is somewhat disconcerting to see him act like it's 100% certain and not simply speculating. Regardless, his explaination is feasable and wouldn't surprise me in the slightest.

 

[Voltron]

Charlie tends to be one of the more reliable Inq folks. Though it is somewhat disconcerting to see him act like it's 100% certain and not simply speculating. Regardless, his explaination is feasable and wouldn't surprise me in the slightest.

Reliable in that he SEEMS to have a relatively high technical understanding compared to most reporters. But you are right - what are his opinions and speculations are really not distinct from objective facts in his writing. It's The Inq so it doesn't really matter, but because of the in my mind he is no better than Fudo. If I had to guess I'd say he just thinks he's a little more important in this world than he actually is.

 

[Razor1]

Voltron, are ya excited about the Transformers movie , j/k, anyways yeah the Inq will get a Fudo replacement

 

[ben6]

..a bit more careful reading would help...

What applications are having trouble with 16x AA on 8800GTX or GTS in terms of performance? I generally use 16x if I'm comfortable with 4x in a game. And apart from the apples to apples performance comparisons doesn't 16x IMPROVE IQ versus 4x or worst case scenario drop down to 4x mode?

 

[Rangers]

I think ATI delayed R600 just for the hell of it.

That really honestly seems to be the case, I'm not kidding.

All indications are they could have launched in January if they wanted too.

 

[Coz]

Charlie tends to be one of the more reliable Inq folks.

I think that's true but he most certainly had his Fudo hat on when he said this....

By delaying the R600 from March to May, ATI suffered a big black eye, but the numbers on the bottom line won't really matter much. The defections should be minimal, Nvidia still does not have a functional G80 driver out, so the pressure is not all there.

Now, some folks may be able to argue a case for there not being a functional G80 Vista driver but he didn't even mention Vista. I'll admit that the G80 WinXP drivers (I'm running 97.92) need a lot of work still (overclocking, dual-view) but to call them non-functional demonstrates breathtaking ignorance.

 

[Natoma]

I think ATI delayed R600 just for the hell of it.

That really honestly seems to be the case, I'm not kidding.

All indications are they could have launched in January if they wanted too.

Ladies and gentlemen, we have tomorrow's Inquirer headline

 

[3dilettante]

If the Inquirer turns out at least partially true, ATI planned a very soft launch (for technical reasons?) and AMD said no.

It sounds more like ATI was really straining with its schedule for R600.
The last-minute cancellations seem to indicate somebody got wise in AMD corporate.

Charlie very usefully (for those who hate to make definitive statements) made it sound like the "entirely marketing-driven" decision could also have a technical reason.

 

[KimB]

If the Inquirer turns out at least partially true, ATI planned a very soft launch (for technical reasons?) and AMD said no.

It sounds more like ATI was really straining with its schedule for R600.
The last-minute cancellations seem to indicate somebody got wise in AMD corporate.

Charlie very usefully (for those who hate to make definitive statements) made it sound like the "entirely marketing-driven" decision could also have a technical reason.

Or the other way around, and AMD corporate finally bowed to pressure by ATI to do it their way.

 

[nonamer]

You guys make it sound like the R600 was always going to (hard) launch in May.

 

[Ailuros]

What applications are having trouble with 16x AA on 8800GTX or GTS in terms of performance? I generally use 16x if I'm comfortable with 4x in a game. And apart from the apples to apples performance comparisons doesn't 16x IMPROVE IQ versus 4x or worst case scenario drop down to 4x mode?

Ben you didn't get my point. NV claims that said issue will transpire to XP drivers also, meaning for one that it's not a Vista only problem and as a second that some sort of "miscommunication" between the driver and some applications occurs so that instead of getting 4x MSAA in some cases 16x CSAA gets enabled instead.

16x CSAA quality truly lies slightly above 4x MSAA and underneath 8x MSAA (always considering Transparency AA too). Apart from the slight performance drop, many won't even notice in real time LOL

 

[Ailuros]

If the Inquirer turns out at least partially true, ATI planned a very soft launch (for technical reasons?) and AMD said no.

It sounds more like ATI was really straining with its schedule for R600.
The last-minute cancellations seem to indicate somebody got wise in AMD corporate.

Charlie very usefully (for those who hate to make definitive statements) made it sound like the "entirely marketing-driven" decision could also have a technical reason.

I honestly hope (given the weird rumours increasing lately) that the explanation is really that simple.

 

[ben6]

Ben you didn't get my point. NV claims that said issue will transpire to XP drivers also, meaning for one that it's not a Vista only problem and as a second that some sort of "miscommunication" between the driver and some applications occurs so that instead of getting 4x MSAA in some cases 16x CSAA gets enabled instead.

16x CSAA quality truly lies slightly above 4x MSAA and underneath 8x MSAA (always considering Transparency AA too). Apart from the slight performance drop, many won't even notice in real time LOL

And as a GTX owner I would rather keep 16x CSAA enabled than have to "Force 4x MSAA" in an application that I won't notice the performance difference in anyway unless I'm running a benchmark script. Rather than call it a driver bug I'd almost NV call it a Feature.

 

[Arty]

I'm beginning to think we need midrange G8x and RV6xx threads.

DailyTech seems to think nothing significant from Nvidia until at least April: http://www.dailytech.com/NVIDIA+will+not+be+present+at+CeBIT/article6200.htm

We already have a thread going on G8x, would you open for RV6xx?

 

[Ailuros]

Now, some folks may be able to argue a case for there not being a functional G80 Vista driver but he didn't even mention Vista. I'll admit that the G80 WinXP drivers (I'm running 97.92) need a lot of work still (overclocking, dual-view) but to call them non-functional demonstrates breathtaking ignorance.

Haven't touched dual-view since I don't need it, yet overclocking works fine once you use Rivatuner instead of that bugfest called nTune.

 

[pjbliverpool]

I think that's true but he most certainly had his Fudo hat on when he said this....

Now, some folks may be able to argue a case for there not being a functional G80 Vista driver but he didn't even mention Vista. I'll admit that the G80 WinXP drivers (I'm running 97.92) need a lot of work still (overclocking, dual-view) but to call them non-functional demonstrates breathtaking ignorance.

I don't think the G80 Vista driver can be called non functional, its working perfectly fine for me on Vista32 and an 8800GTS. Its WHQL certified, purevideo works fine, games work fine, dual view works fine and speed/stability seem, well... fine!

Stability is my only slight concern as I do get a bit of crashing but its generally in modded games so I can't hold nvidia completely to blame. Plus its generally in menus or as the game quits, not in gameplay and doesn't effect windows - I simply restart the app.

I doubt casual users would have any idea at all that there are supposed problems with nvidia's G80 drivers in Vista.

 

[KimB]

The only problem I have with the Vista driver is that the control panel option for flat panel scaling doesn't work. But since the 9x.xx G80 drivers didn't even have a flat panel scaling option at all, at least we know that nVidia is working on putting that option back into the drivers.

I can't wait until I'll finally be able to play my old games that don't support widescreen resolutions so that they aren't stretched. That's been my one significant beef with the G80 so far...

 

[Suflex]

Hmmm... For a part that was supposedly to be released end of this month, R600 info/leaks are surprisingly scarce.

 

[Andrew Lauritzen]

The only problem I have with the Vista driver is that the control panel option for flat panel scaling doesn't work.

Agreed! However since it never worked properly for me in XP x64 anyways, I'm not exactly weeping about it. Still, it would certainly be nice to play non-widescreen-supporting games without stretching. The real annoyance is that many new games have the issue too: it's quite surprising to me how many games do not properly compute a FOV and aspect ratio...

 

[pjbliverpool]

Agreed! However since it never worked properly for me in XP x64 anyways, I'm not exactly weeping about it. Still, it would certainly be nice to play non-widescreen-supporting games without stretching. The real annoyance is that many new games have the issue too: it's quite surprising to me how many games do not properly compute a FOV and aspect ratio...

Games for Windows will hopefully put an end to that