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

The top part of this post details identical speeds and properties between versions tested in 2009. Go to the bottom of this post for actual differences discovered later.

There have been rumblings among some folks in recent months that the PlayStation 3 version of Super Street Fighter II HD Remix is glitched. Having participated in quite a few HDR tourneys on PS3, I’ve noticed nothing different from the Xbox 360 version. I’ve been skeptical of these claims but since I couldn’t say for certain, I decided to test the PS3 and 360 versions out.

First, I would like to note the environment and equipment used. I had 2 monitors side by side: a CRT TV and the Asus LCD monitor used at Evo. For the PS3, I used a 60GB model and for the 360, I used a Jasper Elite. I connected the systems to the CRT through the official composite cables and connected the systems to the LCD via the Elite’s default HDMI cable. I performed control tests by switching TVs and the results were identical so there was no (or undetectably minute) lag. I controlled the PS3 through a HRAP2SA and the 360 through a TE along with official controllers for player 2. I could only do timing tests manually but since 99 seconds was plenty of time for any difference on the 2 monitors to manifest, that wasn’t an issue.

Now, here were my findings after testing both systems under default settings (I didn’t test any other settings) in local versus mode. For each finding, I tested multiple times using different characters and different stages:

  1. The speed on both systems is absolutely identical at every speed level.
  2. The PS3 loading time between rounds is faster by a fraction of a second so it always loads the next round just a bit faster than the 360. This means after the first round, the PS3 is ahead of the 360 but by a constant amount until the next round. After 2 rounds, the PS3 is consistently a few ticks ahead on the game timer.
  3. Stage doesn’t affect the timer speed (nor has it ever affected it in any version of ST). Brazil and Japan (Ryu) end at the exact same time as Spain and India.
  4. Stage doesn’t affect the game speed or length. Constantly jumping with the same character throughout any stage resulted in the same bounce. Also, using Dhalsim to push Zangief from one corner to the other takes the same amount of time in every stage.
  5. In-game slowdown doesn’t affect the game timer. Using lag-causing specials such as projectiles followed by laggy normals didn’t make the timer go slower or faster.
  6. I could consistently reversal out of Ryu’s c.HP on wakeup in either version.
  7. I could consistently produce a 3-hit rekka ken chain in either version.
  8. All of the above regarding HD Remix mode also applies to Classic Arcade mode (ST mode). In addition, HDR’s speed is identical to classic’s speed.
  9. The difference between T1-T2, T2-T3, and T3-T4 is always ~3 timer ticks apiece. The difference between T0 and T1 is a whopping ~16 timer ticks.

So my verdict is that there’s no difference between the 2 versions except that the PS3 has faster loading between rounds.

Now, given that the PS3 and 360 versions appeared to be identical during gameplay, I decided to test out the Dreamcast version of ST (or rather, Super Street Fighter II X for Matching Service), on which HDR was supposedly based. I hooked the DC to the CRT through composite cables and left my PS3 on the LCD. I used an official Agetec arcade stick. And just to make sure I cover everything, I was using a legitimate Japanese copy of X played on a US DC loaded up through the DC-X boot disc. Here were my findings:

  1. Stage doesn’t affect the timer speed (nor has it ever affected it in any version of ST). Brazil and Japan (Ryu) end at the exact same time as Spain and India.
  2. Stage doesn’t affect the game speed or length. Constantly jumping with the same character throughout any stage resulted in the same bounce. Also, using Dhalsim to push Zangief from one corner to the other takes the same amount of time in every stage. Someone here had actually mentioned this before as a difference between the arcade version and the DC version and I can now confirm that.
  3. Even HDR’s max speed of T4 (1 above default) is still a couple ticks slower than the DC version’s T3 (default). And the DC version still has turbo speeds up to T6.
  4. HDR T3 (default) is a couple of ticks slower than DC T2 (1 below default) but a tick faster than DC T1.
  5. HDR is a few ticks slower than DC ST at the same turbo level. HDR T3 is slower than DC ST T3, same for T2, same for T1. This is true whether the DC version is set to fixed or free speeds.
  6. However, HDR T0 is actually the exact same speed as DC ST T0. Different characters, different stages, and induced slowdown made no difference. It seems if the developers did a speed comparison between the DC version and HDR, that was only checked on T0.

My conclusion here (and what most ST players have long commented on) is that HDR is quite a bit slower than the DC version of X. Plus, when you consider that T3 in Japanese X is equivalent to T2 in US ST and that US arcade tourneys have always been one level faster than that (at ST T3), HDR is really slow in comparison. I haven’t checked any of the arcade-to-DC issues yet so I can’t comment on any other aspect of ST accuracy.

Later on, I tried comparing the arcade version of ST to HDR. I used a Sigma AV6000 supergun with the default Sigma 8000TB arcade stick hooked up to a CRT with an S-Video output. On the CPS2 end, I used a US (blue) A board with a US (blue) ST B board. My ST was phoenix’ed but I was using the default US configuration and I left the speed at the default fixed turbo 2. The results were so variant that I couldn’t continue after 2 stages:

  1. In China, the difference in time was consecutively ~5 timer ticks faster compared to HDR T3 and ~3 ticks faster compared to HDR T4.
  2. In U.S.S.R., the difference in time was consecutively ~20 timer ticks slower compared to HDR T3 and ~22 ticks slower compared to HDR T4.

I also did some rough tests in U.S.A. (Ken) and found the results similar to those in China. Remember that HDR stage speeds are constant and note that T.Akiba found China 4th fastest and USSR 5th fastest out of 16 stages in his tests on the JP version (T.Akiba’s SF2 Data). Since I’m not sure how to proceed with this disparity of results, I didn’t.

A basic timeline of DC ST’s speed compared to HDR’s speed goes like this (where each “-” represents roughly 1 timer tick difference by the time a round finishes):

HDR T0=DC T0 - - - - - - - - - - - - - - - - HDR T1 - - - HDR T2 - - DC T1 - HDR T3 - - DC T2 - HDR T4 - - DC T3

Hopefully, these tests have shown that the PS3 version is not glitched as has been claimed (beyond the controller issues that also exist in the 360 version and other bugs outside the battles themselves) or even different from the 360 version. The only possibility of a difference that remains untested is input lag. It also shows HDR’s classic mode does not accurately reflect DC ST’s default speed.

At this point, I don’t think it’s worth my checking for more routine input issues and attributing it to the PS3 platform. If folks will recall, similar “imagining” of issues was present when the PS Street Fighter Collection version of ST was played and again when the DC version was used.

If there are very specific circumstances someone saw that produced obvious results, I’ll still test. For simple combo mess-ups, I can only attribute that to unfamiliarity with PS3 controllers, muscle memory from inaccurate online speeds, or just plain human error. I welcome other folks to test for themselves though.

Differences brought up and corroborated by other folks:

  1. The PS3 version has higher pitched sound effects and voices than the 360 version, esp. noticeable in Chun Li’s voices (credit to skankin garbage).

  2. The 360 version has 2-3 frames of additional input lag compared to CPS2 ST while the PS3 version has 4 frames of additional input lag compared to CPS2 ST (credit to papasi).

Nice job dude. I guess it was just the placebo effect coming into play for most of us.

Thank you for taking the time to do these comparisons.

So if I understand correctly Classic Mode needs to be set to Turbo 4 if it wants to be close to the US arcade free select 3? And even then it will be significantly slower? Cause if I remember right the DC version was still slower than both the arcade X and Turbo.

Well, it’s closer, but it’s still far from close. The difference between HDR T3 and DC X T3 (equivalent to US ST T2) about 6 timer ticks.

First, since zass already proved in his PS2-DC ST tests that fixed and free speeds are for all intents and purposes identical and I’ve come to the same conclusions, I’m assuming that as well (since the DC version, going by JP speeds, doesn’t have a free select US T3). The difference between HDR T4 and DC X T4 (equivalent to ST T3) is about 7 timer ticks, which is a noticeable change.

So going from that I guess the real question now is whether or not Turbo 4 should be the standard from now on? Everything else seems to be good.

HDR has always seemed slow to me. I vote Turbo 4 for purely selfish reasons.

If people are wondering why a reversal against Ryu’s crouching fierce and a 3-hit rekka chain were specifically tested here is most likely why. Sirlin wrote that there were clearly input issues during some of his matches when he was using Fei Long and that, due to input issues within the game, he wasn’t able to consistently use Rekka punches. John Choi and others also claimed that reversals were failing to come out and that it was without question the fault of HDR.

I’ve had clear instances where I’ve been blocking attacks (holding back on the joystick), offline, and my opponent’s attacks were successful even though I was absolutely holding block. I still don’t see how this proves that on the PS3 (and possibly the 360) that HDR isn’t dropping inputs. We’ve seen the problem where you could simultaneously hold forward and backward using the analog and D-pads that had to be patched. It’s going to take incredibly thorough amounts of testing to absolve HDR of its input issues.

One thing if you do have a way of testing that I’d like to know is how many frames of input delay are there in Classic Mode and HDR. If you could test the input delay for HDR that would be pretty excellent.


If you want to play ST then play ST on GGPO or get a CPS2 board and have your friends over. Making HDR as close to ST as possible for the tournament standard is getting old at this point. HDR is not ST and never will be. It’s a new game and should be treated with its own set of rules.

I understand that’s it’s neat to see the differences in speeds between arcade ST, DC, CCC2, and now HDR. Or to test input delay in a game. Threads like this are always interesting and exciting to me. But please, let’s hold off on changing the tournament standard of HDR, merely so that it falls more in line with ST.

Default for HDR should be 16:9, Remixed everything, Turbo 3 (or default) speed. I thought it was lame enough when the EVO rules switched from 16:9 to 4:3 but changing the speeds as well? Enough. HDR is not ST.

If we want EVO to play a game that’s not ST but as close to ST as possible why not fire up AE and allow only ST and Old characters?

Quoted for truth!

I never understood why evo went to 4:3. I thought that was terrible. Is there a good reason for that?

Evo should be run at the default settings of the game.

Is NKI a good reason?

(I’ll let you live that down one day NKI, I really have no beef - it’s just fun blaming you for it. :smile:)

Agreed aside from game breaking stuff like Akuma, etc…

Hm, reading that thread it seems Ponder decided, not NKI. He says

I have to say that as a Claw and Chun player (two of the characters Ponder mentions), I much prefer 16:9. And all HDR tournaments in the northwest have been run at 16:9 (Northwest Majors, for example).

So my vote is for 16:9 at next Evo.

VF4: Dropping inputs is always a possibility. However, what possible operation could there be in the programming that would cause such a thing? The analog-digital glitch was a result of a new feature combined with a programming omission.

If you use a autofire switch on LP to perform LP chains, there’s not a single instance where the LPs are cut off by a dropped input. Try it; even the slow setting on the TE (probably around 10 Hz, if that) never yields a single miss. Now, there’s still a possibility of a particular situation causing an issue but if you have an HRAP, leave LP on full auto (which is set to 23 Hz), mess around with the other side, and see if the LP chain ever pauses when you’re not hitting. I never saw it make a pause while I was messing around. It’s a logical impossibility for me to prove every single case so if anybody can find a reproducible input glitch, that would be helpful.

As for input delay, I’d be very interested in seeing what it is but I’m afraid right now, I don’t have the equipment to accurately count it. Hopefully, somebody else has the will and means.

Regarding tournament standards, HDR will always be at T3 speed with default dip switches, Udon sprites, and OC ReMix music. However, I still think that 4:3 is the optimal aspect ratio but I’ll leave that beaten horse alone and trust ponder to maintain his wise decision next year.

As for using AE, its speed is also off, old characters can’t be throw-softened, and o.Sagat and claw have issues, amongst other differences. And I doubt Evo would be willing to use the JP PS2 Best version of HSF2 for some fixes. But again, this topic wasn’t about ST vs. HDR but rather a look into whether HDR’s classic mode has noticeable differences from ST and primarily a comparison of the 2 HDR versions.

I’m with you there. I also vote for 16:9.

Here’s a scenario. Let’s say that there is some odd programming error where if the 2P controller presses backwards that the 1P controller might accidentally be unable to also press backwards. That would mean that there would be a chance, that if the 2P was doing an attack and holding backwards simultaneously, that the 1P side might not be able to block. This is a crazy hypothetical example but things like the overlapping analog and D-pad inputs give HDR a reputation for the most unique issues to say the least.

Sirlin said that changing things around had tons of unintended consequences in the balancing stages. I couldn’t find the entry on his site but he talked about how he changed Honda’s buttslam hitboxes around and he started noticing that he was getting hit by attacks from the across the screen for no reason. So he turned on the hitboxes and did a butt slam, and sure enough, his changes had inadvertently added in stray hitboxes to the buttslam move.

We know that Zanief’s SUPER command had several input issues that were resolved with the most recent patch. However T.Hawk has the same issues with his new SUPER command and they still remain in the game. It is possible when you have a full SUPER meter with T.Hawk to perform the new motion, end the joystick in back or downback, and get a dragon punch. This happens from within and out of throw range for Hawk in HDR.

I don’t know what Sirlin or Choi did to get the game to drop their inputs but replicating the scenario is the best way to test it. Who knows, maybe Ryu can’t reversal DP against a meaty cr.fierce only on certain stages, or only from the 2P side? Or maybe Rekkas drop against only Claw and Cammy. It would require so much time and effort to test these out that it almost seems not worth it unfortunately.

Here is a 16:9 vs 4:3 poll with 60% of people favoring wide over standard.

Anyways if this a comparison thread we might as well get into the 16:9 and 4:3 thing as well in my opinion.

NKI was the most outspoken player for using 4:3. I think the reasons NKI came up with for using 4:3 are what Ponder used to make his decision.

This is what Damdai had to say about it.

Well, I think we should really push to have it to 16:9 for next Evo. I completely agree with what Damdai wrote.

Thanks for doing all the testing Ganelon! I’m not sure why the two versions felt different to me. Maybe I’m just crazy.

And yeah, I hope Evo goes widescreen next year.

good shit, i always knew hdr felt way slower.

Thanks Ganelon for the tests. Speed 3 definitely felt slower the moment I played Remix for the 1st time. That’s why I always play training mode and vs mode offline on speed 4, cuz it feels right.

You know, this is a pretty weird difference, but having played PS3 and XBox360 versions of this game, there is one difference…all the sound effects on the PS3 version are slightly higher pitched, maybe even a bit faster, but I don’t think so. Unfortunately, the only reason I notice this is because I have perfect pitch, so I don’t have any empirical data to support my claim.

I also prefer 16:9. 4:3 just looks really ghetto, like straight from the hood. Anyways, thanks for the testing Ganelon.