The fault with DF was not recognizing the problem with their footage. That, or recognizing the footage was screwy and still publishing it that way.
Digital Foundry Article Technical Discussion Archive [2013]
- Thread starter helio
- Start date
-
- Tags
- digital foundry
- Status
The fault with DF was not recognizing the problem with their footage. That, or recognizing the footage was screwy and still publishing it that way.
It is not like I never criticize them but like everybody I'm sure they are facing "hard" timelines, they got to publish something. Not too mention that if it was the final release and they had the game for them-selves they could have done the captures again (both the ps4 and XO).
They are certainly faulty in some ways though business is business, the press is not apart of other businesses in this regard.
We have to consider that they had to complete these in a short timeframe. Granted they should have caught the problem and notified the readers that there indeed is a problem, but at the root of it it really isn't DF's fault that something such as RGB limited/full is broken on one set of hardware (which nobody ever expected) and apparently nobody from Dice nor Microsoft bothered to inform them about it.
As a side note I do appreciate, even though the results were messed up, that they insist on using the same tool chain they are using usually for the sake of consistency and trying to hold true to their "best practices". It did not paid this time around, shit happens I guess 
If they were so pressed for time then why did they waste it by not using the same setup as everyone else?
Deontology? Or they have great tool chain which usually does not fail them (actually it did not)?
They remain the reference at their job, imo they tried to maintain their quality standards, it failed.
Now with other sites posting comparisons and not having the game and systems to redo the capture I think they have to release what they had, standards only stand that long in the face of missed clicks
Current iPhone/iPad is roughly half the GFlops, in a more flexibly programmable GPU with better bandwidth/ALU ratio while using an extremely bandwidth efficient architecture.
Next year will see a transition to 20nm (which is where iOS devices will exceed PS360 graphics capabilities), and the year after that (perhaps more significantly) a transition to 16nm FinFET.
Last edited by a moderator:
Probably because the default setup wasn't up to standards.
If they didn't use their own equipment, we probably wouldn't have gotten the frame rate test.
But on the other hand way more OS overhead?
Ignoring the libellous nature of this post, if MS demanded equality, why not just render the PS4 version at 720p and keep it exactly the same as XB1? And how come the XB1 version isn't the definitive one if that's what MS paid for?
Crazy talk, quite frankly. Maybe you were joking?
I'm sorry but this is silly. Their whole point is grabbing the best possible footage, which means Full RGB, 1080p, and 60fps. Their whole point is not having footage in the first place, but for being able to do a technical analysis, and this is what their requirements are and should be.
Everyone saying that the footage was screwy and that they should have checked is misguided - this is not about getting the best possible picture out there because you are a broadcasting company and it's all about the viewer's experience, this is about getting the best possible capture for technical analysis. As many said before, 30fps is completely unacceptable (and the rate that everyone else got) as the framerate analysis component far outweighs any coloring issues. Whether you think it is more important to compare limited range output just because the Xbox One had an issue at this point, or whether it would have been more important to take the original footage is up for debate. Everyone else has the limited range footage, so that wouldn't have brought much to the table, imho. Instead now we know that the Xbox One has an issue, and can look for Microsoft to clarify that the issue is fixed at some point before launch, or whenever.
For me it definitely matters - I get a clear picture improvement when I select Full RGB, and for those using PC displays for their console it would matter too, as their displays generally only work with 0-255 output in the first place.
Michellstar
Regular
Correct
Add to that
"EA provided Elgato Game Capture HD consumer-level devices at the Stockholm event. They're great little units but they are unsuitable for DF work because they cannot acquire 1080p at 60 frames per second and they do not provide lossless video"
What kind of compression is using Elgato?
Michellstar
Regular
I don´t know, i´ll have to check
But people are complaining about DF face-off in relation to the other publications which are using as they said a lossy compression.
I presume DF captures with a lossless codec up to 60fps and 1080p, yeah they got wrong the colour with X1 and try to explain why.
A pity that BF4 was out before nextgen consoles were on streets, in reality they are comparing pre final builds with dev-kits
TheWretched
Regular
But if the Renderer -> HDMI is at fault, why is it, that all the other footage on the web doesn't have this issue? Does HDMI provide color range sensing (I mean, maybe the elgatos do say they're limited RGB, the HDMI says it's limited to the renderer and it's alright... and the other way around the HDMI port still uses limited, but reports full range?)
Is there any analysis of other footage if it's limited or full range RGB?
Is there any analysis of other footage if it's limited or full range RGB?
As I had mentioned multiple times, the X1 build is very similar to the PC build, while the PS4 features different assets at times. This is actually visible in the histogram above. where X1 follows similar curvature to the PC, except the black portion is completely crushed, see how the PC curve have another 1/16 that's not shown on the X1.
Other footage on the web were probably captured with limited range, 30fps that EA provided. If other people recorded limited range it's not a surprise that they didn't get hit with the same issues, as DF had much better control over the equipment they used.
If DF deliberately used different settings for the three setups, there would be merit in saying DF screwed up. The fact that they used the same settings and one console in particular is producing problems tells us DF is probably doing their job correctly.
If anything I give them credit for following protocols and workflows strictly.
These tests should be done with a little tweaking as possible to reveal issues.
As I had mentioned multiple times, the X1 build is very similar to the PC build, while the PS4 features different assets at times. This is actually visible in the histogram above. where X1 follows similar curvature to the PC, except the black portion is completely crushed, see how the PC curve have another 1/16 that's not shown on the X1.
Do you have a picture of the source of the three histograms?
It doesn't look to me as if they're of the same scene.
TheWretched
Regular
The thing is... if the console puts out limited and the recorder is set to full, it's less of a deal than the other way around, since you only lose precision, but not information (i.e. you can stretch it afterwards). What was it?
Exactly.
DF captured Full range, so under the circumstance that everything is working properly, having crushed blacks is illogical, pointing to an issue with either the XO hardware/software.
The following is a possible scenario.
Which is pretty screwed up in my opinion.
Sure. My point is that the X1 histogram lacked the lower end of the PC.
http://www.eurogamer.net/articles/2013-10-29-battlefield-4-next-gen-vs-pc-comparison-gallery
used 46 to 48
- Status
Similar threads
