first r420 review leak

Evildeus said:
Well, it's not my word it's the words of the speaker. Now why should i take your word, because it's yours isn't it?

Look... you said the speaker said, so it's your word until the speaker pops in and says so himself, right?

But, like I said, that's fine with me. Even more so because there were two "14k"s, one on the presentation and one in a preview (FiringSquad, I think) that did use 800x600. Since I wasn't there at the presentation thus couldn't hear the speaker himself, and have not been told of 14k beeing a legit default score due to OC so far, I (falsely) concluded that the two "14k"s are the same. BZB asked, I answered, you answered otherwise. I just asked for evidence because I had none supporting the "14k with OC" theory and when you responded along the lines of "well, the presentation said so", I specifically noted that I believe you even though that's not exactly direct evidence. No need to look for personal attacks where there are none.
 
Unknown Soldier said:
http://www.chip.de/ii/21775269_b35653becd.jpg

So what is the yellow connector for??

Looks like an analogue audio connector to me - Maybe part of the video encoding capabilities is some kind of audio encoding support?
 
volt said:
To all idiots editing / deleting and messing up Bjorn3D articles: grow the f*ck up.
To all idiots caching and exchanging our NDA'ed material: grow the f*ck up.
To all idiots posting links to our NDA'ed material on different websites: grow the f*ck up.

Sorry for the OT but some of you need to grow the f*ck up, literally :rolleyes:

Good call volt. WTF did everyone's morals fly out the window???

I expected more than this from B3D posters. Goes to show how things have changed around here since I started visiting.
 
I think the little yellow connector is video in. Yellow is typically used for video connections, and the cards do have a Rage Theater chip on them which would indicate VIVO capabillity.
 
ace.jpg
 
JCLW said:
I think the little yellow connector is video in. Yellow is typically used for video connections, and the cards do have a Rage Theater chip on them which would indicate VIVO capabillity.

From [H]:
"The yellow connector on the rear of the card is for a composite video-in connector so that you can run a cable to your video capture port on the front of your case if you have one. ATI says this will most likely not be on the shipping cards in North America as it is a more popular feature in Europe."
 
Small comment from Bjorn at bjorn3d.com.

The aceeditor that were linked to isn't our 'real' editor. It is/was a goofing-around editor that was set up to allow staff to try out a WYSIWYG editor that I converted from ASP to PHP. No live reviews were in there and staff were using it to see if it would help them write reviews quicker.

I screwed up by not protecting it. Since it wasn't linked and it didn't contain any live stuff, the protecting was on my list of things-to-do after all our server-issues were resolved (we had to move the site twice in the last month - last time to a new dedicated server) and I just hadn't gotten around to that. The x800 preview was in that editor for ... 2 days.

I'm amazed someone took the time to rootscan us for unprotected pages. I'm also amazed they picked the time while our x800 article was there.

It's bad enough someone was willing to copy the text we wrote, it's even worse that they decided to change it. All that anti-NVIDIA crap is pure BS. We aren't more anti-NVIDIA than we are Anti-ATI.

That said, the x800 XT is a sweet card. I just wish I had been able to keep it longer than the 4-5 hours I had to benchmark :(.

/B
 
Ailuros said:
Ok I just received an indirect warning considering that specific slide.
Beyond3D Review said:
The basic pixel output engines remain unchanged from R300. In normal operation each pixel pipeline can output 1 colour value and 1 Z/Stencil value, but there is no optimised Z/Stencil rendering performance.

Sooo... it's back to 16*0 for R420 and 8*0 for R3x0, rrright? :?:
 
Joe DeFuria said:
anaqer said:
Sooo... it's back to 16*0 for R420 and 8*0 for R3x0, rrright? :?:

Correct.

Is anandtech right in saying that with fsaa it acts as a 32x0 though
http://www.anandtech.com/video/showdoc.html?i=2044&p=5

By contrast, R420 pushes 32 z/stencil operations per clock cycle when antialiasing is enabled (one z/stencil operation can be completed per clock at the end of each pixel pipeline, and one z/stencil operation can be completed inside the multisample AA unit).

The different approaches these architectures take mean that each will excel in different ways when dealing with z or stencil data. Under R420, z/stencil speed will be maximized when antialiasing is enabled and will only see 16 z/stencil operations per clock under non-antialiased rendering. NV40 will achieve maximum z/stencil performance when a z/stencil only pass is performed regardless of the state of antialiasing.
 
Is this Z-while-AA trick specific to the R420, or does this exist in R3x0 too (but somehow just haven't been talked about yet)?
My money is on "R420 only", but a boy can dream. 8)
 
anaqer said:
Is this Z-while-AA trick specific to the R420, or does this exist in R3x0 too (but somehow just haven't been talked about yet)?
My money is on "R420 only", but a boy can dream. 8)

I'm pretty sure it applies to R3xx also. Too lazy to check B3D's past reviews to find out. ;)

That being said, being able to perform 2 "z-operations" per clock is not necessarily the same as performing 2 z-writes (or 2 z reads) per clock. So I'm hesitant to say that when in AA mode, the R4xx is on par with the NV4x in terms of stencil only pass rendering. I believe NV4x will still have an advantage here, but someone else will have to verify.
 
Back
Top