NVIDIA shows signs ... [2008 - 2017]

Status
Not open for further replies.
link it I was under the impression gddr5 uses less power per clock then gddr3 and it is considerably less.

GDDR5 and GDDR3 aren't meant to run at the same frequency. What would ahppen is if you try and hit a certain bandwidth, you could use a N wide GDDR3 interface or a N/2 GDDR5 interface. That saves you power, die area, etc.

The problem is that NV isn't keeping bandwidth the same, they want to increase bandwidth and get better performance. That costs more power, especially if they have a wider memory interface than ATI (and it seems like that is quite likely).

really that was very profound, performance still drops with ATi's tesselator, its around 30% performance hit when going to heavy scenes. Also stream out performance is very important as well.

https://a248.e.akamai.net/f/674/920...sselationForDetailedAnimatedCrowds_SLIDES.pdf

The point of tessellation is to substantially increase scene quality. If you just use tessellation naively, it does reduce performance.

The point was that if you use it correctly, you can get 2-3 orders of magnitude more polygons, and you reduce memory bandwidth. It does cost you, as you say, 30% performance. But ALU performance is cheap. Bandwidth is not and 3-4 orders of magnitude more polygons can be noticeable in some cases I would imagine.

DK
 
I was under the impression that they discontinued 4830 and 4850 while making the 4770.
As I know, former plan was to terminate production of HD4830/4850 in March. But this plan was changed and its production was extended by 2 months (at least). I think ATi won't stop it until RV740 will have decent yields.
 
GDDR5 and GDDR3 aren't meant to run at the same frequency. What would ahppen is if you try and hit a certain bandwidth, you could use a N wide GDDR3 interface or a N/2 GDDR5 interface. That saves you power, die area, etc.

The problem is that NV isn't keeping bandwidth the same, they want to increase bandwidth and get better performance. That costs more power, especially if they have a wider memory interface than ATI (and it seems like that is quite likely).


yeah but it would be nice to get clarification from charlie's point of view.

The point of tessellation is to substantially increase scene quality. If you just use tessellation naively, it does reduce performance.

The point was that if you use it correctly, you can get 2-3 orders of magnitude more polygons, and you reduce memory bandwidth. It does cost you, as you say, 30% performance. But ALU performance is cheap. Bandwidth is not and 3-4 orders of magnitude more polygons can be noticeable in some cases I would imagine.


not disagreeing with that. Just stated it had a performance hit.
 
Yeah, actually I didn't undelete anything, Chrome screwed up. And the post that was deleted wasn't the one I thought, instead it was one which said "You know that rant earlier when I accused someone on this forum of willfull ignorance and stupidity. Pat yourself on the head, you are now the champ there." - I've seen worse, but given that it did bait Razor who flamed in response, it didn't add anything to the thread. And since Charlie is gone, why bother...

Your saying you let me retain the crown :) Thanks man that is swell. Hate it when an usurper sweeps and and swipes that thing.
 
That story isn't on the Inq now, so given the number of timezones Taiwan is ahead of the UK, can we guess it didn't happen? I can't believe Nvidia would hang a sign out like that - it would all be done quietly behind the scenes.

It took a year for the AMD/ATI merger, and we didn't hear any rumours till the very end. Likewise, all we got a couple of years ago about the stalled talks between Nvidia and Intel were vague rumours after the fact.
 
Well screenshots are also easier to photoshop than actual links to an article. ;)

Not saying those are photoshops, just that it's quite possible they could be.

Regards,
SB
 
I can confirm that was on their site earlier. It's missing now so they pulled it off it seems.

I was doing my early morning tech reads on all sites, Inq and Fudo first (since I am in North America and they the first to post ) and read that story and its only when it was mentioned that someone could not see it that I checked again and voila-- gone .
 
Forgive my ignorance but how can you put yourself up for sale when you dont technically own yourself....nvidia became a listed public company 10 years ago i think. Shouldn't they be the property of the various shareholders now?

Maybe they mean nvidia has received a takeover offer for the shareholders to consider? That would be some news. Makes quite a change from hearing about governments rescuing/taking over private firms that have got in trouble.

yeah but it would be nice to get clarification from charlie's point of view.
If he doesn't return to this thread I suggest you try his new site here: http://www.semiaccurate.com/

From the 3 articles posted there appears to be at computex at the moment...if you are in taipei might be able to get answers personally this week.
 
I saw the story too before it was pulled.

Didn't a similar thing happen about the AMD-ATI take-over ... a story appeared on the Inq very early in the proceedings and got pulled rapidly?
 
Well, now. I wonder why he left.
Well he was pretty much the only original member left from before Mike sold the site. Theo, Fuad, and Mike have all gone on to found new sites, so he might as well also.
 
Last edited by a moderator:
Status
Not open for further replies.
Back
Top