133 replies to this topic

[Dimento Graven]

Der Geisterbaer, on 01 February 2019 - 11:18 AM, said:

...

Hint: Videos are not much of a proof.

OLD videos aren't, correct, HOWEVER, new vids obviously not edited? Very much proof...

[Der Geisterbaer]

Nightbird, on 01 February 2019 - 11:25 AM, said:

Anecdotal evidence the best evidence x)

Sigh ... there's no "anecdotal evidence" involved once you are provided with the raw data that would allow you to replicate the result. That's actually how peer review normally works.

But for the fun of it: Provide me with an easy to access upload space and I'll gladly give you some footage

[Verilligo]

Der Geisterbaer, on 01 February 2019 - 11:18 AM, said:

I gave you all the necessary numbers on my macro usage to do a reasonable peer review as any serious proof would actually demand. You suggested that I do private lobby testing. Now how about you doing that yourself with my numbers instead of going your well established fallacy route?

Hint: Videos are not much of a proof.

I disagree entirely, videos are tremendous proof. In fact, it's the only way of reliably gathering data, given you MUST record from the target's view and your own view to actually see what each of your heat bars are doing in relation to one another. If you had a means of recording the target's current heat level accurately without recording their view, then certainly you could just post the raw data. But since you can't, you need a recording from their point of view. Once you have that, it's not TOO difficult of a proposal to simply combine the two videos together.

It's impossible to do a good peer review without access to the results of the original test to compare your fit, though. We don't even have enough information about your experimental set up to mirror your test conditions, let alone replicate your results, whatever those results may have been. How many HS do you have? What were you attacking? How many HS did they have? Did their ping matter?

[Dimento Graven]

Der Geisterbaer, on 01 February 2019 - 11:46 AM, said:

Sigh ... there's no "anecdotal evidence" involved once you are provided with the raw data that would allow you to replicate the result. That's actually how peer review normally works.

But for the fun of it: Provide me with an easy to access upload space and I'll gladly give you some footage

Not that difficult to save to youtube.com, or if you like Google Drive, put it in a folder, create a sharable link and we can all download it.

[Verilligo]

Dimento Graven, on 01 February 2019 - 11:48 AM, said:

Not that difficult to save to youtube.com, or if you like Google Drive, put it in a folder, create a sharable link and we can all download it.

Might also be able to use gfycat if they don't like Google for some reason.

[Der Geisterbaer]

Dimento Graven, on 01 February 2019 - 11:48 AM, said:

Not that difficult to save to youtube.com, or if you like Google Drive, put it in a folder, create a sharable link and we can all download it.

I do not have either a youtube account or a google drive account and I certainly do not intend to go through the trouble of registering with either service just to satisfy the unwarranted demands of people that have no clue about how "proof" and "peer review" (or statistics for that matter) actually work.

[Dimento Graven]

Der Geisterbaer, on 01 February 2019 - 11:52 AM, said:

I do not have either a youtube account or a google drive account and I certainly do not intend to go through the trouble of registering with either service just to satisfy the unwarranted demands of people that have no clue about how "proof" and "peer review" (or statistics for that matter) actually work.

Huh... Well... That's an interesting attitude to take. Registering for either is free and easy.

Uploading to YouTube is only slightly more involved in that YouTube will go through a re-render process and you have to set the viewability of the vid.

I've uploaded... SEVERAL vids myself, many being recordings for proofs of bugs in game.

[Der Geisterbaer]

Verilligo, on 01 February 2019 - 11:48 AM, said:

I disagree entirely, videos are tremendous proof.

No they aren't.

Verilligo, on 01 February 2019 - 11:48 AM, said:

In fact, it's the only way of reliably gathering data,

Let's just better agree to disagree on whether or not data gathering could be done in a more reliable way.

Verilligo, on 01 February 2019 - 11:48 AM, said:

given you MUST record from the target's view and your own view to actually see what each of your heat bars are doing in relation to one another. If you had a means of recording the target's current heat level accurately without recording their view, then certainly you could just post the raw data.

Within reasonable margin of error I could simply record voice com and have the target mech's player announce current heat levels. But this would just bring forth accusations of fudging as if this particular demonstration would actually require a level of scientific proof that real problems would. Not to mention the fact, that with the data I gave you could replicate my test much faster than ...

Verilligo, on 01 February 2019 - 11:48 AM, said:

But since you can't, you need a recording from their point of view.

... I could ever set up such dual recording or ...

Verilligo, on 01 February 2019 - 11:48 AM, said:

Once you have that, it's not TOO difficult of a proposal to simply combine the two videos together.

... post-processing the footage in a video that meets your expectations of convenience.

Verilligo, on 01 February 2019 - 11:48 AM, said:

It's impossible to do a good peer review without access to the results of the original test to compare your fit, though.

All a matter of "margin of error" and "satisfactory degree of reproduction". We're not talking measurements on gravitational waves here but ultimately a rather simplistic time measurement under macro conditions. Even if you can't match my initital environment (computer with all its specs including peripheral as well as latency to gaming servers) you should be able to replicate the observed behavior close enough if you really wanted to engage in peer review there.

Verilligo, on 01 February 2019 - 11:48 AM, said:

We don't even have enough information about your experimental set up to mirror your test conditions, let alone replicate your results, whatever those results may have been.

You want more detailed info on my setup? I can give you that far easier and faster than I could ever provide the desired video.

• Mech: PIR-CI
  
• Loadout-String: AE:D40C0|Xe|n<2|LCp@0|l^|iB|iB|iB|UR|jB|jBq@0|l^|iB|iB|iB|UR|l^|jB|jBr<0s<0t@0|l^u@0|l^v40|iRw504040
  
• Relevant Skill nodes: a total of 3 Heat Gen nodes and one flamer ventilation point, nothing else that would affect flamer effectiveness
  
• Macro functionality: Logitech Gaming-Software Environment with LUA-scripted macro that abides to the following rules: If mouse button 4 is pressed then loop while mouse button remains pressed. During the loop press 5 (weapon group that has been assigned with the four flamers) for 14ms followed by 36ms of not pressing any button)
  
• Map: Tourmaline desert
  
• Mode: private lobby with a friend

Verilligo, on 01 February 2019 - 11:48 AM, said:

How many HS do you have? What were you attacking? How many HS did they have? Did their ping matter?

Let me put it this way: None of these questions would have been answered to a satisfactory degree by the video that you think would be "good proof" beyond you knowing that a Highlander IIC was involved.

Edited by Der Geisterbaer, 01 February 2019 - 12:27 PM.

[Kubernetes]

thievingmagpi, on 01 February 2019 - 09:35 AM, said:

Monday as per Russ' twitter

See y'all on Monday then.

[Verilligo]

Der Geisterbaer, on 01 February 2019 - 12:22 PM, said:

Let's just better agree to disagree on whether or not data gathering could be done in a more reliable way.

That's not how data gathering works. There is always a way of getting better, more accurate data. It's just a matter of how much effort you're willing to put into it and how much return you get on the extra effort invested. I get what you're trying to say, that investing more effort to get data that's within a half second of certainty is unnecessary. But you're the one making the accusation that flamers are broken with macros, which is a problem that was supposedly taken care of before. That means the burden of proof is on you. Surely if you gave enough of a damn to do testing on this and record it, you'd give enough of a damn to post the recording despite the minor inconvenience.

Quote

Within reasonable margin of error I could simply record voice com and have the target mech's player announce current heat levels. But this would just bring forth accusations of fudging as if this particular demonstration would actually require a level of scientific proof that real problems would.

I'm not going to accuse you or your buddy of outright lying. But the difference of 2-4s in this particular example is a HUGE percentage jump, despite being sufficiently small that someone could have just counted time badly. That's a thing that people do, we have metronomes specifically because a lot of people are garbage at counting proper seconds, myself included.

Quote

You want more detailed info on my setup? I can give you that far easier and faster than I could ever provide the desired video.

• Mech: PIR-CI
  
• Loadout-String: AE:D40C0|Xe|n<2|LCp@0|l^|iB|iB|iB|UR|jB|jBq@0|l^|iB|iB|iB|UR|l^|jB|jBr<0s<0t@0|l^u@0|l^v40|iRw504040
  
• Relevant Skill nodes: a total of 3 Heat Gen nodes and one flamer ventilation point, nothing else that would affect flamer effectiveness
  
• Macro functionality: Logitech Gaming-Software Environment with LUA-scripted macro that abides to the following rules: If mouse button 4 is pressed then loop while mouse button remains pressed. During the loop press 5 (weapon group that has been assigned with the four flamers) for 14ms followed by 36ms of not pressing any button)
  
• Map: Tourmaline desert
  
• Mode: private lobby with a friend

Let me put it this way: None of these questions would have been answered to a satisfactory degree by the video that you think would be "good proof" beyond you knowing that a Highlander IIC was involved.

That's a lot of very good information, thank you, though it would have helped to know details on the Highlander IIC because their own cooling rate would impact this. Would also be nice to know how many seconds it took you to max out their heat at the 90% mark in each test.

[Der Geisterbaer]

Verilligo, on 01 February 2019 - 01:27 PM, said:

That's not how data gathering works.

Datagathering works via measuring. Video footage can be used to do such measuring but it isn't necessarily a good source. Video footage in this particular situation does not directly provide data with the kind of quality that you claim it would have. The very necessity of how I would have to rig the recording and then have to post-process the footage goes counter to claim of that being "good proof" because it has too many systematic inaccuracies and still is prone to manipulation.

Verilligo, on 01 February 2019 - 01:27 PM, said:

There is always a way of getting better, more accurate data.

Indeed and video footage on its own is not very accurate.

Verilligo, on 01 February 2019 - 01:27 PM, said:

It's just a matter of how much effort you're willing to put into it and how much return you get on the extra effort invested.

And here's the fun part: I already have spent enough effort into getting the data that interested me personally.

Verilligo, on 01 February 2019 - 01:27 PM, said:

I get what you're trying to say, that investing more effort to get data that's within a half second of certainty is unnecessary.

In a 15 vs. 30 seconds till overheat scenario pretty much everything you demanded is unnecessary and unwarranted.

Verilligo, on 01 February 2019 - 01:27 PM, said:

But you're the one making the accusation that flamers are broken with macros, which is a problem that was supposedly taken care of before.

I'm not even making that particular "accusation". I'm merely attesting to personal experience in my tests that do no coincide with said fixing of a problem or Nightbirds claim that his own private lobby tests proved that this has indeed been fixed.

Verilligo, on 01 February 2019 - 01:27 PM, said:

That means the burden of proof is on you.

Only as much as I do burden myself with that particular proof and only up to the point where noone who is truly interested cannot - under any circumstances - replicate my results.

Verilligo, on 01 February 2019 - 01:27 PM, said:

Surely if you gave enough of a damn to do testing on this and record it, you'd give enough of a damn to post the recording despite the minor inconvenience.

And here's the big misconception: The test itself was super easy to conduct because I (last night) had all necessary ingredients already available: a running Logitech macro environment and a suitable macro that needed only little number tweaking. Commenting in this thread also took up virtually no time or ressources. Recording a second go of the tests today took slightly more time and ressources. Uploading those results on the other hand will take even more ressources and inconvenience me in other areas that are none of your business, so that's not going to happen unless someone provides a solution that accommodates my conveniences.

Verilligo, on 01 February 2019 - 01:27 PM, said:

I'm not going to accuse you or your buddy of outright lying.

I actually don't care what you intend to accuse me of or not.

Verilligo, on 01 February 2019 - 01:27 PM, said:

That's a lot of very good information, thank you, though it would have helped to know details on the Highlander IIC because their own cooling rate would impact this.

Actually the Highlander's cooling rate is most likely a negliable variable since it's the same under either firing condition and would only make the heat transfer in staggered flamers worse but never better.

Verilligo, on 01 February 2019 - 01:27 PM, said:

Would also be nice to know how many seconds it took you to max out their heat at the 90% mark in each test.

I didn't give you these numbers for a reason but try to have a guess where I got my "estimation" of 30% increase of time till hitting 90% mark. I'll even say that - with closer inspection - it could have been a 50% increase til heat saturation on the target which is still significantly lower than the close to 100% time increase until overheat shutdown that I experienced on my end.

Edited by Der Geisterbaer, 01 February 2019 - 02:12 PM.

[MechWarrior5152251]

I completed both events (Not Solaris), did not see any RAC2 that I can recall. If they were a big deal I would think I would have noticed one in the 40+ matches I played this week?

I did notice a lot of focus fire on me when I was using an LB2X, maybe they thought I had a RAC2 and all went after me?

[Phoenix 72]

Well, I did my first match today since the RAC exploit became public knowledge. And had one guy on the red team that shot way too many RACs way too fast for way too long to seem normal to me. And one of my favourite Mechs is the Triple RAC Bushwacker. Our Assaults folded like nothing at the time, too. Almost stopped playing right there.

But decided to see whether this persisted. Got another 3 matches in with nothing. However, on the next three matches I typed in all chat that using macro's with RACs was considered a bug exploit and might result in bans before the game. I still lost 3 out of the 4 matches that I played, but that was due to normal stuff.

So my experience right now is 1 in 4 matches. Not sure how widespread it is. Will play some more later.

***EDIT: 5 or 6 games later, no new sighting. Seems like people are getting that they are not supposed to use this...***

Edited by Darakor Stormwind, 01 February 2019 - 09:01 PM.

[Mister Glitchdragon]

My first match tonight was a RAC-Boat/PIR hellscape. Glad I stuck it out; the rest of the night went okay.

[Maddermax]

Played a few games yesterday, didn’t notice any major issues with it or 5xRAC type builds, but it’s possible I just wasn’t a target for them. I did feel sorely tempted to take a Piranha using the MG exploit after trying it on the testing grounds, but decided against it in the end - Just not my style and it wouldn’t be right, but I understand the temptation. I did try it on my new stock purifier (2MGs) in a match, as I had heard contradicting info about it working/not working for MGs in live matches, but after firing and hearing the sound bug that meant it was working, turned it off again.

Hopefully they address the MG and flamer issues at the same time as the RACs.

Edited by Maddermax, 02 February 2019 - 02:01 AM.

[justcallme A S H]

RACRO is being hotfixed on Monday according to Russ twitter.

Wait and see!

[Insignus]

justcallme A S H, on 31 January 2019 - 02:10 PM, said:

I mean you claimed the Match Maker let Tier 1s in with Tier 5s not two days ago and it's like that from 2017 and I should talk to Chris about it

Devil's Advocate: The tiering system does let people into tier 1 who BELONG in Tier 5

[Dee Eight]

Insignus, on 02 February 2019 - 10:51 AM, said:

Devil's Advocate: The tiering system does let people into tier 1 who BELONG in Tier 5

Tier is NOTHING EXCEPT AN XP bar for time spent playing the game. Players who have been at the game long enough eventually reach Tier 1 regardless of how unskilled / unintelligent they actually are.

[Gantry]

Racro has got to go. Ever since that youtube video posted, it's rampant, and it's ruining the game!

[Orville Righteous]

Gantry, on 02 February 2019 - 01:47 PM, said:

Racro has got to go. Ever since that youtube video posted, it's rampant, and it's ruining the game!

I guess I'm lucky because I haven't seen it. If I did see someone who I thought was using it, I'd just report them and move on. I've enjoyed the matches I've had since I found out about the exploit. I think the vast majority of the community isn't exploiting.