Trying to figure out how inconsistent the load times really are between systems. Currently we assume they're inconsistent, but it would be nice to actually test and prove it.

Take part of the experiment:

1. Download the save file. Ask for the save file from my Discord if it's not working. Indrek#9137 https://www.dropbox.com/s/asqjw5mx92c6wfc/Helene.d2s?dl=0
2. Add -seed1400148573 to your target line to keep Jail spawn consistent.

Start the timer. Use Telekinesis on waypoint, walk towards the champions, cast one nova, save & exit. Repeat until level 16 and stop the timer.

My submission: My first slowest attempt was RTA: 5:53 / IGT: 4:20 Time difference: ~93 seconds

After this I did a lot more attempts and time difference was always the same ~93 seconds. Here's my fastest attempt: RTA: 5:42 / IGT: 4:07 Time difference ~93 seconds

Bonus experiment: -ns comparison RTA: 5:13. While IGT saved 90 seconds, then using -ns only saved 30 seconds.

Probably won't be the last experiment. For example should also test shrine checking load time consistency.

Here is one attempt (without -seed) I did so far RTA 5:26 / IGT 3:46 which gives ~99-100 seconds

@BokserKabaty I think you also didn't have -seed active. Your spawns seem inconsistent. Better to keep -seed for now so we have some consistency.

Edit: Actually it's probably best to test both anyway. I'll do some runs without -seed as well and see if I get similar results.

I did some runs without -seed and also got ~99 second difference in RTA/IGT. I'll post some videos and start a Google sheet to keep track of all submissions.

@Indrek You're right, I just forgot to restart a client for -seed to apply. I can feel difference between the two. It's also very promising you've got similar load time results in real world scenario w/o -seed.

Here is one with -seed applied: RTA 5:51/IGT 4:16 ~95seconds, so pretty good. I also had two 6min attempts, with 1-2 more loads = diff was about ~98 seconds.

Based on these facts, we really have to recalculate the time in the old world record so that everything is fair

Could someone refer me to a runner who is known to have bad load times? I looked at top sorc runs and didn't see anything out of the ordinary there when it came to load times.

Egg plays on a chitty laptop, maybe take a look at his runs?

I would like to, do you have any links, to see his runs ?

old vods from egg got deleted during the dmca stuff

Just wanna throw this in here - might be relevant info I spend a little time looking into slimos jail leveling and the loading screens with -ns seem to vary, depending on how much time he spends within a run

He does a total of 69 Jail runs to get to level 18 20-21 of them were with Frost nova - they took around 1 Second of load time total 48-49 with nova - These runs had load time ranges from 1 - 3 seconds

While using frost nova and in the first couple of levels with nova, the loading screens were shorter - but he was spending more time within the jail The more damage nova did, the longer the loading times were

Quick (konservative) math suggests: 20 * 1 seconds = 20 Second time loss compared to our ingame timer 49 * 2.5 seconds = 122.5 Second time loss compared to our ingame timer

Adding up to 2 mins and 22.5 Seconds If i subtract this from his total time - that run barely stands as WR there - not accounting for loading screens outside of the leveling proccess, which I ASSUME cant be more then 30 seconds Therefore I would estimate slimos runs to be somewhere between 56:30 to 57:00 when put into IGT

Looking at the average time difference from IGT vs RTA on the jail runs - we are looking at around 4 minutes

Extrapolating that to the RTA time with Sound would be above 1 hour

DISCLAIMER: This was a very "scratch the surface" analysis - I think I will go into more detail on this some time soon, as the "counting runs" method is a bit sketchy

Current WR in HC P1 Sorc is 1:00:41 (IGT). Yesterday I got a 1:02:00 (IGT). The run in IGT is 1 minute slower, but when comparing RTA the run is almost over 3 minutes faster.

1:00:41 (IGT) - 1:05:12 (RTA) 4min 41s difference 1:02:00 (IGT) - 1:02:45 (RTA) 45s difference

The run yesterday got a leveling setup where I didn't need to do as many save and exits as usual so my IGT/RTA time difference ended up very small. This is the reason I couldn't make act 2 leveling work anymore as well, because IGT hugely favors strategies that save & exit more.