The new prototype driver doesn't fix AMD's hardware issue where one 17 frame old frames stack?
Who are you and who do you work for?
Who are you and who do you work for?
It doesn't fix the 2 scanline issue... Let's just not talk about that :ninja:
So you have zero integrity and you expect us to believe all this other crap you're shoveling as well?
Whatever hardware issues AMD has isn't a fault of the program, the prototype drivers fix the real issues (RUNTS and Frame Times) enough for me. It's not as good as nVidia but I never expected it to be, thankfully I wasted only a bit of my life arguing with you over a total non-issue.
Whatever hardware issues AMD has isn't a fault of the program, the prototype drivers fix the real issues (RUNTS and Frame Times) enough for me. It's not as good as nVidia but I never expected it to be, thankfully I wasted only a bit of my life arguing with you over a total non-issue.
What I told you has been proven not to be crap, why can't you accept that you're talking to someone who actually has access to this stuff.
The hardware issues aren't the fault of FCAT, they're just interpreted wrong by FCAT. Yes, the prototype drivers fix all the real issues.
No it's crap and non consequential, as I said on probably page 2. I also haven't seen anything of the sort where it has any real effect on the graphs.
It's a strawman you used to attempt to discredit a good program, nothing more.
The only thing we learned from this thread is you have no integrity whatsoever with zero disclosure, or of course you're a liar, pick your poison I guess.
It's a strawman you used to attempt to discredit a good program, nothing more.
The only thing we learned from this thread is you have no integrity whatsoever with zero disclosure, or of course you're a liar, pick your poison I guess.
Would that be the prototype they sent out with the 7990 release, or the tweaked near beta release one :sneaky:
Sushi, I dont mind about CF performances, but are we seeing some new AMD gpus soon? I want price wars 
u know something?
u know something?Last I checked, wrong was wrong, and a tool which wants to be taken as reputable and having integrity needs to give accurate results.
A hardware vendor which wants to be taken serious needs to produce products that aren't broken.
Last I checked, a tiny line at the border of your monitor every few hundred frames isn't "broken"?
Black Octagon
Golden Member
- Dec 10, 2012
- 1,410
- 2
- 81
Eh? And you are who exactly? Don't believe you've disclosed this so far in the thread...might be called for since you seem to be suggesting (if not screaming) that FCAT is tarnished by a conflict of interest due to it being an NVIDIA-designed tool
The issue is Nvidia knows about the issue (on AMD cards), Nvidia "tried" to fix the issue, therefore they know about it, but they never bothered actually fixing it when their solution didn't work.
No it's broken, it's just inconsequential.
And you're opinion is that FCAT is a bad program because AMD's hardware spits out a 17 frame old RUNT every few hundred frames..
What a joke, I guess it comes with a lack of integrity. :thumbsup:
AMD's hardware doesn't appear to clear the top 2 lines of the frame buffer every once in a while. It's not 17 frames old, and it isn't a runt. It's like this
FRAME1
FRAME2
---- NEW FRAME
2 scanlines of FRAME1 show up
FRAME3
FRAME4
etc.
It's invisible to the naked eye.
I don't know why I've spent this much time of my life explaining this to as ungrateful of a moron as yourself.
It's a runt, but I'm not even sure why we're still talking.
You "supposedly" broke your NDA to further your position in an argument on a forum, but won't say who you are or who you work for.
Nothing you say has any merit at this point, not only will you not self disclose, but you've already broken an NDA... Nice character reference. :thumbsup:'
The more likely case though is you aren't under NDA and you're just lying, which is just sad.
You "supposedly" broke your NDA to further your position in an argument on a forum, but won't say who you are or who you work for.
Nothing you say has any merit at this point, not only will you not self disclose, but you've already broken an NDA... Nice character reference. :thumbsup:'
The more likely case though is you aren't under NDA and you're just lying, which is just sad.
It's not a runt, a runt is a frame in the proper order but a small size in relation to the frames before and after (~10%). This is an artifact if you will.
I'm not breaking anything to further my position in an argument, I'm letting people know about issues with FCAT. You are just arguing it because you're, frankly, an ungrateful imbecile who can't accept things being shown right in front of his face.
Yes of course I am lying, given than SKYMTL confirmed what I said and that article I linked to confirmed the exact same 2 scanline issue.
Go troll somewhere else.
Jaydip
Diamond Member
- Mar 29, 2010
- 3,691
- 21
- 81
Looks like a runt, walks like a runt, talks like a runt, it's probably a runt. Isn't that your issue? Nvidia is calling it a runt?
What else are we going to do with something that shouldn't be there? Ignore it? If it's in the recycled trash that doesn't belong there it's still being displayed. What do you do with it? Nvidia tried to help AMD out and ignore it, but they couldn't fix AMD's hardware issue in their output charts. Luckily it's not a huge problem, and has little to no effect on the actual results.
You claim nVidia is inflating the frame rate with reported FPS charts, but compared to FRAPS they're actually under reporting, even on their own hardware.
You said you were under NDA and leaked information about the hardware issue for AMD not being fixed which is under NDA. That's breaking the NDA :thumbsup:
An issue with AMD's hardware, not FCAT.
Last edited:
No, a runt is a frame in the proper order. This frame isn't in the proper order, it's an artifact.
You could ignore it, consider it 1 runt (very reasonable), but what SHOULDN'T be done is calling it 15 missed frames. Nvidia could have easily fixed it, they know that they didn't. "Actual results" aren't the whole story and FCAT is giving an incorrect picture of the problem.
You're using a test that we don't have full knowledge of the procedure to claim that. Evidently, FCAT can't do a proper job of reporting what FRAPS would display, so they shouldn't display that at all. It's incorrect. Just show actual results.
So AMD's hardware is faulty and artifacts at default settings?
Can I RMA my cards and claim they can never properly replace them with non defective hardware?
I'm just glad you went with such a silly strawman argument. Grats on breaking your "NDA", I hope it was worth it.
Can I RMA my cards and claim they can never properly replace them with non defective hardware?
I'm just glad you went with such a silly strawman argument. Grats on breaking your "NDA", I hope it was worth it.
BrightCandle
Diamond Member
- Mar 15, 2007
- 4,762
- 0
- 76
No its not a runt, its a separate issue.
AMD's cards don't scan out the frames in order properly. Its kind of incredible to say that actually but they don't. Instead sometimes the prior frame leaks through, now it could in theory have been a runt frame that leaks through and never gets shown otherwise, so there are a couple of unique cases. If that leak through is slightly to large for the detection algorithm then it will call it a runt and not ignore it as a scanning out error. In that case there is every possibility that the software will now assume 15 missing frames and jump the FPS. The actual FPS rather than the fraps estimated one will still be accurate, but if you use its FPS without ignoring runts data then it could be wrong. It would basically be reporting a lot more dropped frames than there actually were, because of this leak of scanlines.
I think this is a genuinely plausible problem in the analysis part of the software of FCAT.
This is honestly a big problem for AMD's hardware, it makes some dreadful tearing when it happens, people see it on the desktop all the time at eyefinity resolutions and so far AMD has said nothing on it, and its another problem FCAT has shown up. Analysing from just the frame data is quite tricky.
AMD's cards don't scan out the frames in order properly. Its kind of incredible to say that actually but they don't. Instead sometimes the prior frame leaks through, now it could in theory have been a runt frame that leaks through and never gets shown otherwise, so there are a couple of unique cases. If that leak through is slightly to large for the detection algorithm then it will call it a runt and not ignore it as a scanning out error. In that case there is every possibility that the software will now assume 15 missing frames and jump the FPS. The actual FPS rather than the fraps estimated one will still be accurate, but if you use its FPS without ignoring runts data then it could be wrong. It would basically be reporting a lot more dropped frames than there actually were, because of this leak of scanlines.
I think this is a genuinely plausible problem in the analysis part of the software of FCAT.
This is honestly a big problem for AMD's hardware, it makes some dreadful tearing when it happens, people see it on the desktop all the time at eyefinity resolutions and so far AMD has said nothing on it, and its another problem FCAT has shown up. Analysing from just the frame data is quite tricky.
Yeah I think we've pretty much arrived there BC, thanks for the post!
Nothing wrong with FCAT, it was a strawman argument against it. The problem is actually AMD's hardware, and the slight jumps of FPS aren't even the real issue, just an inaccurate attempt to mimic FRAPS readouts. The important part, the filtered results are unaffected.
Seems FCAT has exposed a lot of problems in AMD's hardware, perhaps AMD should thank them!
Nothing wrong with FCAT, it was a strawman argument against it. The problem is actually AMD's hardware, and the slight jumps of FPS aren't even the real issue, just an inaccurate attempt to mimic FRAPS readouts. The important part, the filtered results are unaffected.
Seems FCAT has exposed a lot of problems in AMD's hardware, perhaps AMD should thank them!
Last edited:
Jaydip
Diamond Member
- Mar 29, 2010
- 3,691
- 21
- 81
Ok let me understand this "FRAME1 FRAME2 ---- NEW FRAME 2 scanlines of FRAME1 show up FRAME3 FRAME4 etc." This is what you said.So are these "scanline frames" even smaller than runt frames? does this happen irrespective of any game(i.e. is it a universal thing for amd)? it looks like wasted resource to me to display "scanline frames" .Does fraps detect these frames as being displayed?
BrightCandle
Diamond Member
- Mar 15, 2007
- 4,762
- 0
- 76
TRENDING THREADS
-
Discussion Zen 5 Speculation (EPYC Turin and Strix Point/Granite Ridge - Ryzen 9000)- Started by DisEnchantment
- Replies: 25K
-
Discussion Intel Meteor, Arrow, Lunar & Panther Lakes Discussion Threads
- Started by Tigerick
- Replies: 21K
-
Discussion Intel current and future Lakes & Rapids thread- Started by TheF34RChannel
- Replies: 23K
-
-
Facebook
Twitter