Retroarch
3 years ago
Canada

you may watch this. emulator at its best with runahead isnt even matching the 2 frame input delay of my console. at best it's doing is 3. the worst 6.

emulator starts first

Edited by the author 3 years ago
Washington, USA

How many frames was runahead set to in your test?

Washington, USA

I think it's fine to allow runahead to be used up to 2 frames then. There's no advantage over console but even if there was and you had a 1 frame delay, it still wouldn't help - console already feels lagless. You won't instantly become better at the game. It'll just level the playing field and make emulator less annoying to play on.

Remz likes this
Canada

This is the correct move to allow 2F runahead as there is no benefit over playing on a console, but I am still set that someone just needs to play it and any doubt would be removed.

You would just need the emulator, the approved and supplied snes9x core in the "Leaderboard Rules and Requirements" thread and a ROM.

I am having trouble with understanding why the moderation team is unable or unwilling to do so.

Remz likes this
Washington, USA

I played a little bit with runahead off, on 1 and on 2. There's still input lag on 2. It's noticeably worse than console but it's better than not having it on.

After playing around with different combinations of settings, these were the best that I found without compromising emulator performance https://imgur.com/a/sUeKTc7. With a better computer, you could probably run frame delay at a higher setting and even use hard gpu sync at the same time. That might lower the input lag further but I doubt it would be better than console.

Edited by the author 3 years ago
Canada

I have not used any frame delay. I don't understand what bad effects it might make if set too high so i've left it untouched

Washington, USA

Not sure what would happen gameplay wise at the highest number of frames but if your computer can't handle it, the emulator basically runs in forced slow motion and it becomes unplayable lol.

Edited by the author 3 years ago
Canada

https://www.reddit.com/r/RetroArch/comments/6e5o8z/what_does_frame_delay_do/dibzn1h/

i am reading this, but also dont really understand how it works. I dont know what adverse effects I should look for like obvious frame skipping if runahead is too high.

"The higher the Frame Delay value, the faster the system needs to be to have the frame ready on time"

I would like to imagine my system is above average and well equipped to handle an emulator's toughest settings. What if i dont stutter? I have some learning to do but this might wait until the weekend when im not stuck for time. Any information and test results I perform put will go here.

Canada

My run has been rejected on the grounds of a rule stemming from this discussion, yet made before any moderator was able to demonstrate that one would be a better player with the setting turned on. They have still been unable to prove so.

Even after one moderator has given their playtest results and siding towards my claim, the evidence him and I have presented has been disregarded, essentially being told my numbers are wrong and my hardware is faulty.

All of this being orchestrated by a single top level moderator who refuses to acknowledge new information or even actively investigate speedrun related matters.

Behaviour like this is unfair, unethical and detrimental to the integrity of discussions when they are forced to be one sided and even when other moderator input is ignored. Why is this tolerated? Where are the other users who have allegedly voted in favor of the ban, and who audits the audit?

Brooklyn should be commended for being the sole person willing to investigate and I want it known that all he had to do was play for a short time to be able to feel the *still present disadvangage in input lag on emulator.

I am not afraid of being shot down. But i am asking it be done fairly.

https://www.speedrun.com/yi/run/mr47lw8z

United States

hi My concern here is that we don't know where the discrepancy between your test and Brunnis' test comes from. Brunnis seems to believe that people often experience higher latency on emulator because of problems with hardware or setup (as you can see in their final paragraph). So my question is whether the difference is because or some error on Brunnis' part or if there really is a loss of 2+ frames from something in your hardware (perhaps the adapter, although 2 frames seems extreme). I'm not really sure how to test this other than to buy the exact same hardware they used and copy the experiment.

Also to be fair there was a discussion on this in the Yoshi discord, and in an informal vote 8 people voted unanimously to ban runahead, so this wasn't a solo decision by mt even if it seemed like it on here. However, I do think we should be willing to reconsider if there's evidence that Brunnis' test is somehow incorrect.

Canada

What exactly is it going to take for someone to OPEN THE EMULATOR, as asking 7 times doesn't do it? Idk yo, test it on your individual setups. I'm not the one with a sword on my name yet I've run and shared my multiple tests. Do you need more? I can do more.

God forbid you actually test it. One step at a time. You can feel the lag

Canada

I thought i would level the playing field to something closer to Brunnis's hardware and perform the same test i have before on my desktop. the adapter i own is also a raphnet adapter. i would expect it to perform reasonably close to the one Brunnis uses. With a sandy bridge Thinkpad:

3F 5F 5F 4F 5F 4F 4F 4F

the average input latency being 4.25 frames. an increase of .25 from the 4 frames average of my desktop.

setup including jar to balance my phone: https://i.imgur.com/Pzjo9rV.jpg

Does anyone else wish to add to the sample size of 2 sets of hardware?

North Carolina, USA

[quote]My run has been rejected on the grounds of a rule stemming from this discussion, yet made before any moderator was able to demonstrate that one would be a better player with the setting turned on. They have still been unable to prove so.[/quote]A possible advantage has already been demonstrated, and explained to you in detail, multiple times. I very clearly laid out the logic and you haven't engaged with any of it. The only thing you brought up was a critique of the methodology, which I subsequently addressed and showed it did not meaningfully impact the findings.

[quote]Even after one moderator has given their playtest results and siding towards my claim, the evidence him and I have presented has been disregarded, essentially being told my numbers are wrong and my hardware is faulty.[/quote]Nobody has said your numbers are 'wrong.' Your hardware setup could introduce additional lag, this is true. As I've said though, it's ultimately irrelevant.

[quote]All of this being orchestrated by a single top level moderator who refuses to acknowledge new information or even actively investigate speedrun related matters.[/quote]I have explained multiple times that 'new information' doesn't impact this decision. Go back and read the logic that was very clearly laid out for you.

[quote]Behaviour like this is unfair, unethical and detrimental to the integrity of discussions when they are forced to be one sided and even when other moderator input is ignored. Why is this tolerated? Where are the other users who have allegedly voted in favor of the ban, and who audits the audit?[/quote]Nothing unfair or unethical has happened and further accusations of these or similar things will get you nowhere. Also, you voluntarily left the discord where these things are discussed. Those in favor of the proposal: shado, crispy, thurler, me, gill, forte, icterus, suus. Those against: none.

[quote]God forbid you actually test it. One step at a time. You can feel the lag[/quote] Feelings are not sufficient to override empirical data. You don't seem to understand this.

Here is the simple version: It has been shown that runahead can result in an advantage over console. If an emulator or emulator feature can result in an advantage over console, it is not allowed. Therefore, runahead is not allowed.

If you can engage with any of the reasoning I've laid out that's perfectly fine. Any additional 'just open the emulator' or 'unfair and unethical behavior' posts are going to be deleted.

Edited by the author 3 years ago