The thing is that unlike security cameras, the game does not tell you how many data each mission has, and if someone tries this run, there is a chance this player might believe to have collected everything, but forgot a few data because of not knowing that certain areas have to be investigated even with the surveillance class' passive ability active. From what I know, for U.S.S., there are 17 in ch.1, 38 in ch.2, 29 in ch.3, 26 in ch.4, 19 in ch.5, 17 in ch.6, and 9 in ch.7, but who knows if I missed a certain corner despite playing U.S.S. campaign for years. For Echo Six, I currently don't know and a detailed investigation would be necessary.

I may try test runs of that category for both campaigns when I come back to this game to see if something else needs to be considered as well.

In the rules, it says "Obtain all collectibles (cameras, datapads, raccoons, typewriters) and finish the game." I have a few questions because some aspects are not fully clear to me:

• Are all data within a mission meant, or just enough to obtain S-Rank in that sub-category? While most missions have slightly more than the required amount of data for the rank, U.S.S. mission 2 (City Hall) is the extreme case where you need 20 for S-Rank, but there are actually a total of 38 data available in this mission.

• If you die, all data collected before the checkpoint respawn, but the number of collected data does not reset back to the number collected at that checkpoint. Therefore, it is possible to have more data than the maximum available by death and respawn. Or in other words, you could farm the required amount of data in a good spot. What are the rules about this circumstance ?

• You can only interact with Raccoons and Typewriters for the first time. Once collected, the command to interact with these objects is blocked (There is this red circle you also see when you try to pick up ammo or grenades despite having max capacity instead of the button). Is it enough to just get near them and make the (blocked) command appear ?

It was a little hard to find something about the history of this game, but according to some findings, PS3 received the patch 1.03 that made major changes to the gameplay (I have checked my PS3 game and can confirm that it runs on 1.03). I couldn't find a source stating what version XBOX 360 is on, but due to observing similar bahaviour to the PS3 version, my guess is that it runs on 1.03 as well, but I don't know. The version most different is the PC version and has a complete different version number. It seems to run on 1.2.1803.132 or 1.2.1803.135. Despite 1.2 being a much higher number than 1.03, I didn't observe any of the 1.03 changes of PS3 being in the 1.2 . So I believe the cause of currently discovered differences between PC and PS3 (and 360 ?) and potential future discoveries of differences are these two different version numbers.

When I was trying to figure out how exactly IGT is calculated in the last USS mission 'End of the Line', I came across some inconsistency I don't know where they come from how IGT is determined in PC versions. For Console versions, I got consistent results so far.

This is how the IGT matched with the calculations in most runs. (Being a few seconds off was totally fine for me because a) I subtracted video times and added them together, so slight inaccuracy is expected, especially when doing that about 3 or 4 times , b ) I cannot tell at what precise milliseconds the IGT stops and continues.). I examined both Solo and Coop runs.

• Before climbing the ladder to Leon and triggering the cutscene Prior to the final Encounter, everything plays out as normal (Timer stops during skippable cutscenes (CGI's and the in-game cutscenes where the screen fades out to black), Timer does not stop for unskippable in-game-cutscenes (the ones where the screen does not fade out to black), Deaths do not stop the timer). Death and respawn prior to reaching this point does not appear to have any effect on the final encounter.
• The cutscene with Leon prior to the final encounter is unskippable because of the decision you have to make near the end, but the timer stops during that part.
• However, when you see that cutscene for the first time, the timer continues during the 'Defend, Execute' decision phase and adds this Duration of 12 seconds to the timer.
• Timer runs as normal for the final encounter. In case you wonder, the 18 second introduction sequence where the game tells your objective depending on your choice does count as well.
• If you die and respawn for the first time during the final encounter, the timer does NOT continue and add 12 seconds during the decision sequence a second time. (- Unknown if any changes occur when dying more than once. There is only one run where that happens and this particular run has one of the inconsistency I am going to mention further below.)
• The final cutscene stops the timer. However, if you decide to watch the final cutscene, bear in mind that the timer indeed continues during the credits (at least for the defend Leon route this is the case).

Now here are two kinds of inconsistency (both happened only on PC runs) I have run into and that got me confused while trying to figure out how IGT is calculated on chapter 7. I don't know if I am allowed to mention the exact runs directly, so I am not going to for this post, just to be safe:

• In some PC runs, the 12 seconds were not added during the decision sequence despite seeing the cutscene for the first time. For other PC runs, they were. This saved 12 seconds for those runs.
• In two PC runs (one Solo and one Coop), the IGT appeared to have simply stopped during the final encounter. And in both cases, it stopped at the same moment. After the 18 second introduction sequence of the final encounter, the time calculated up to this point approximately matches the final IGT. For the Coop run, it happened right away, in the Solo run after dying once during that encounter. These two runs had bigger timesaves like 30+ seconds because of this.

I have no idea so far how and why this happens, but I think I should mention the potential presence of these occurences.

@rozukk I guess you refer to the many deaths you had in the final encounter on chapter 7. I don't know any details, but I have noticed that dying once or multiple times during this encounter does something to the IGT that keeps it much lower than it is supposed to. I have seen this on a Let's Play where someone tried to defeat Leon for about 50 minutes and died 8 times until giving up and going for defend instead, but ended up with an IGT of 19:11. I have no idea when, how and why that happens (and if it is restricted to console versions before receiving the patch and the PC version), but this odd occurence is the reason why your mistakes on the final encounter didn't have severre consequences on the time and you still got a good 04:25.

When doing side-by-side comparisons in order to identify points for optimization, and due to the lack of competition in this game overall and especially on the PS3 version, I used runs from different platforms for comparison. This is where I noticed that the IGT on the three platforms does not represent equal performance. After comparing the video intervall between the opening CGI of a stage and the result screen with the in-game-time (assuming cutscenes are skipped immediately and the game was not paused) for multiple runs on each platform (only the recent ones that show the in-game-time could be used though), it appears that while the IGT of the PC version is accurate, the one of the XBOX 360 version is delayed by about 3 seconds and the one of the PS3 version by about 12 seconds. So far, the mentioned numbers are only estimations. The accuracy of the PC version and the delays of the PS3 appear to be stable though. No statement can be made about the XBOX 360 version because only one run exists that shows the in-game-time.

In summary, the IGT's don't show equal performance and the consoles have a delayed IGT, so a distinction between platforms for full runs is not optional. IL runs are only reserved for PC versions since XBOX 360 versions cannot compete on heavily optimized IL's while PS3 versions cannot compete at all and have to consider full runs instead. (All information provided here refer to the 'Solo' category)

I have played the PS3 version and have seen a few videos of the PC version and I believe to have observed the following differences (as far as I know, PS3 version received a patch that the PC Version didn't):

• On PC version, you can use melee as much as you want. On PS3, you get a cooldown after using melee twice in a row (it can be bypassed though. If you don't plan to transition into a brutal kill animation, you can cancel the recovery animation by holding the shooting button. If you hold the shooting button and time pressing the melee button correctly, you can not only melee as much as you want on PS3 as well, but also melee faster in general.).
• On PC version, you can dash as much as you want. On the PS3 version, however, the dash drains all your stamina and takes away the ability to sprint or dash for about 2 or 3 seconds.
• On PS3 version, it is possible to use the dodge to jump into cover. The dodge animation can therefore be canceled preliminary by touching anything that supports cover. I haven't seen this being possible on the PC version. Therefore, the PS3 version enables another trick. The dodge jump seems to instantly refills your stamina because if you jump into cover every now and then while sprinting, you can sprint forever.

I cannot tell to what degree these three aspects can affect speedruns. I also don't know anything about XBOX 360 because I hardly find videos of this version.

About your suggestion to time the speedrun by just adding all six race times, I did further analysis the past days and this is what I found: Timing by race times would actually give me the World Record back. I present the lap times of the current top 3 runs as an explanation. I didn't consider milliseconds in my calculation so far because I don't know how to integrate them in my Software: 1st ; 2nd ; 3rd 03:00:898 ; 02:57:172 ; 03:03:693 02:44:209 ; 02:38:939 ; 02:44:249 02:09:594 ; 02:08:831 ; 02:13:059 03:25:761 ; 03:21:374 ; 03:25:541 03:16:657 ; 03:15:640 ; 03:21:361 04:26:000 ; 04:27:118 ; 04:37:280

19:00:000 ; 18:46:000 ; 19:23:000

Which means 1st place beats 2nd place by 21 sec in RTA but loses about 14 sec in Race Time (RT). 1st place beats 3rd place by 45 sec in RTA but only by about 23 sec in RT. 2nd place beats 3rd place by 21 sec in RTA and even by about 37 sec in RT. 1st place did noticable mistakes in scrolling through menus while 2nd and 3rd did it almost perfectly. To my mind, these differences are definitely worth a discussion in this community. Another thing I did was to compare all RTA and Race times of different regional versions of Chronicle Mode speedruns. As I lack a video for Chronicle Mode on XBOX 360, I took a Season 1 Speedrun instead. The findings are that RTA and RT times are always identical regardless of NTSC or PAL or loading times. So all XBOX users have equal chances in terms of racing performance. However, XBOX 360 has slowdowns on most (not all) tracks, so RTA can be about 8 sec to over 1 min slower than RT. Slower gameplay means more reaction time and therefore better chances. The disadvantage in RTA would turn into an advantage in RT for 360 users. These are my findings so far. Timing via Race Times is worth a thought, but we need more opinions on that.

I cannot fully answer this question, but I can share a few ideas about aspects that might influence the loading times. First of all, I don't think there is a big difference between NTSC and PAL, if at all, in terms of loading times. Based on my observations, loading times for both a race and the parts selection menu are about 6 to 7 seconds for both versions, while for two speed runners, the race loading time is 3 seconds and the menu loading times in general are 2 to 3 seconds. Here are the three ideas which might influence loading times:

• XBOX Hardware Version: As far as I heard, the XBOX has multiple versions (v 1.0 to 1.4 and v 1.6) where different hardware components like disc readers were used. Depending on which version you have, it might improve loading times a little.
• Disc version: There are copies where the two games 'Sega GT 2002' and 'Jet Set Radio Future' are on one disc (together with some demos of other games). The Jet Set Radio Future speedrun community already found out that their game has differences in loading times between the version that contains Sega GT 2002 and the one that doesn't, with the latter being the better one. Something similar might apply the other way round as well, but I can't tell if the difference would be a major or minor one.
• Installed version: I'm not an expert about this topic, but I heard that you can softmod the XBOX and that allows you to install the game data from the disk to the hard drive and launch the game from there. The rules for another game that I speedrun banned copies installed on the hard drive because load times were heavily improved. About the XBOX 360, I have seen a speedrun of Career Mode Season 1 done on a 360 (not on this leaderboard). The game suffers from slowdowns on the 360 that vary on each track, so I think players who run on the 360 actually have a disavantge compared to XBOX users. These are all ideas I can provide. I hope any of this helps.

No major skips have been found so far. Two 'Out-Of-Bounds' were discovered by a player, but both cause the character to move very slowly which is apparently the game's reaction to not knowing if the character is currently in a 2D or 3D area. So this is probably not suitable as a speedrun strategy. There are few minor tricks which are specifically useful for speedrun purposes, but most of them are limited to individual levels or individual level runs. I can create a list with some techniques in the guides if the mod(s) are okay with that.

I have a question about the Full Game Any% category. The rules state that 'Any save file is fine, but you must complete every level in order for the run to be valid'. From what I read from this thread, this rule was primarily created for the possibility to skip cutscenes. But already created savefiles can have another advantage we should be aware of. If you have all stars and if you don't improve your scores on a stage, the game won't update your scores and won't open the save menu. That allows you to scroll to the next level immediately which saves about 4-5 seconds per stage. Multiplied with 17 stages (not counting the last one), this allows you to save about 01:00 to 01:30 minutes, which is a lot for a short game like this one. That means, the best savefile would be a 100% completed savefile which has very good IL times for every stage so it would be unlikely that they were updated during a Full Game run. My question is if making use of this advantage is valid for Full Game runs, or must IL times of the current run be visible which would mean that you have to pick a save file with high IL times so they would be updated for each stage,and thus simulating a new game file ?