Comparison of HDR Versions (PS3, 360, DC, CPS2)

Ha, my opinion means nothing but I wanted to bring it up just because it seemed noticeable to me in that particular situation and in case folks were forgetting about the input lag clause here. Like I said, I favor no console, only accurate facts. A test is still the only thing worth mentioning in the first post.

Also, I highly doubt Evo will change consoles. Like I said, SSFIV will remain PS3 no matter what and it has a much larger audience. If there’s a 1-2 frame delay, we would probably have to deal with it for now just as they are. Plus, it’s too early to mention that when no tests have been performed yet. You can’t make such a huge change without proof.

Anyway, Shari brings up a good point: it would be even more worthwhile to see whether the natural input delay is still 4 frames (plus 0-1 frame input lag in the DC port). It seems the frame recording folks can determine those numbers as well though.

Guessing it’s GGPO ftw when it comes to ST!!!

I believe DGV mentioned some testing he was planning to do to compare offline versions. But if it’s ever proven with concrete evidence that PS3 has more input lag than 360, then I will update the Wiki (or somebody else can do it).

Apparently, PS3 has 1 frame of input lag. I updated the Wiki.

Well it appears that its 1 frame + any that the xbox version has.

That’s for SF4. As far as I’m aware, HDR has yet to be determined, which is why I didn’t bother mentioning anything yet in the first post here.

Okay cool, thanks. I corrected the Wiki.

I was told, at SCR that 4x3 and Widescreen mode have different speeds, but I did some testing, and the frame skip is the same or very close to the same.

So, I finally got around to putting together an input delay testing set-up last night. It looks like Xbox360 HDR is at least a frame slower than Arcade Super Turbo. (This is unsurprising considering it’s based on the Dreamcast version.) It appears that (1) different sorts of input changes are handled differently by the system, and (2) that input delay is affected by turbo frame skip, so I’ll have to work up a testing program. Once I have those pieces, I may track down a PS3 to do version comparisons, or reactivate my XBLA account to test the effects of network smoothing on input delay.

Thanks for starting the testing, Rufus. NKI tested the DC version to have 0-1 frames of input delay so if it’s a range of 2 values, then that should be normal (since every other port varied similarly).

The PS3 version clearly has something wrong. Speed 3 on ps3 is way too slow, and there is general wonkiness of dropped inputs, or something that makes things just not come out when they should. The HD Remix players I’ve played with have all agreed that something is really wrong with PS3, they don’t exactly know what. I have muscle memory to do rekkax3 easily, and I could do it only about 25% of the time on PS3 speed 3, and then 100% of the time on Xbox speed 3, same madcatz TE stick. In any case, the PS3 version should not be used for any kind of competition. Much better to spend the effort to set up xbox 360s, even if players bring them. I also bet a lot of the people with a negative opinion of HD Remix played on PS3 and might not realize it’s screwy.

PS3 speed is exactly the same as 360 speed save for loading times (where PS3 is faster). I’m willing to bet anyone any amount of money on the accuracy of my speed test in the first post. Or like I said, you can just do like I did and compare the 2 versions side to side: start a match with the same characters on both setups, hold up, and notice that the jumping rate is identical and the match lasts the same relative time.

Rather, the difference in versions has almost certainly to do with with input lag instead (although this hasn’t been scientifically tested). A lot of players use visual identifiers for combos in conjunction with muscle memory (which I didn’t even realize until I came across this case) and by the time you already see it on the PS3 version, it’s too late. To confirm that what I’m saying, keep practicing rekkas for an hour on the PS3 version until you get them 100% of the time. Then, immediately switch to 360 and you’ll notice you’re doing them way too early. I guarantee that anybody who tries this test will come to the same conclusion. At least 3 frames input lag is my guess for the PS3 version.

I also felt lag on the 360 version compared to CPS2 but it’s certainly not as major and it’s possible the dynamic stage speeds in the CPS2 version is throwing me off. Hopefully, Rufus’s work will reveal the answers. But either way, I agree with Sirlin and the majority that folks playing on HDR should use the 360 version whenever possible. I’ve bolded my input lag stipulation in the 1st post to make it clearer to everyone and will revise the entire thing once someone uncovers concrete evidence of this or other differences.

I’m pretty sure that EVO is sponsored by Sony, so the Xbox is going to be a very hard sell for anything official.

Direction Test 1
Negative Edge Uppercut Test

it’s probably cause I’m slow but whats the conclusion from those two tests?

well?

The last two posts humor me.

I think it basically means 360 HDR has the same input lag as the DC version (4 or 5 frames). CPS2 ST has 4 frames of input lag for comparison (I think…).

Yeah, that’s about the size of it.

PS2 CCC2 ->CRT 8 frames # [media=youtube]JoJzobmdGzU[/media]
PS1 ST ->CRT 4 frames # [media=youtube]JoJzobmdGzU[/media]
DC ST ->CRT 5 frames # [media=youtube]JoJzobmdGzU[/media]

cps2 s-video encoder->CRT 4.5 frames *

cps2 vga encoder->LCD 7-frames *

xbox classic mode->LCD 6-7 frames *

xbox classic mode->CRT 6-7 frames *

PS3 classic mode->LCD 8 frames

ps2 ae->CRT 6 frames *

ggpo
w7 aero 8-11 frames (and a lot of dropped inputs) *
w7 fullscreen 7-9 frames *
w7 disable aero 7-8-frames *
xp 6 frames *

supercade
w7 aero 8-10 frames (and a lot of dropped inputs) *
w7 fullscreen 7 frames *
w7 disable aero 7 frames *
xp 6 frames *

xp offline fba version 029708 6 frames *

xp offline mame version 0141b 6 frames *

xbox SSF4 ryu cr. rh 6 frames *

PS3 SSF4 ryu cr. rh 7-8 frames (most of them at 8 frames) *

Console hdmi output to Asus V236H in gaming mode

Emulation are tested on Dell Dimension 9200 2ghz core2duo / ATI 4850 / Asus V236H

thanks ultracombo for lending the ps3 for testing