Well, I was planning on writing this to you in private, but haven't gotten around to it yet. In the mean time, it seems you have posted publicly ousting me on various things, so I'd like to give you a view from my (objective) perspective. Hopefully I've addressed everything that you mentioned, but let me know if I missed anything.
Background: I had to remove woc from the dev team a few days ago after I was alerted by a large number of emails that he was deleting all his code from the osu! repository. I did this as a safety measure and still have plans on letting him return should his mind change.
woc posted a long rant on baidu (http://tieba.baidu.com/p/2905827484?pn=1) to which I will provide responses to. Yes, I had to use google translate to retrieve meaning, but I believe that for the purposes at hand that should be fine. I am avoiding any of the emotional language as that is not relevant.
update: better translation in this post. I won't update my responses below as I don't believe it changes too much, but does allow for a much smoother read for those interested.
I accepted that this was my fault for accepting your code changes which caused this.
I was working with that code for an hour or so before realizing I was editing your method rather than the one actually being used. This is very hard to see due to it containing the same 500+ lines. Therefore, I removed your method to avoid future confusion. Tom has already implemented what you were doing using a much nicer solution (as he required similar behaviour for difficulty calculation).
Also you fail to mention the number of times you submitted patches which wouldn't compile (you never tested that it compiled before submitting). Or the recent bug fixes you have submitted which almost always cause a really obvious regression that can only mean they were never tested.
Yes, I can be stubborn when I am confident that my opinion is right, but this is how I choose to run osu!, and it lets me run it in an efficient way.
I'd just like to end this by saying that it is *very* easy to call a person out like this. No one is perfect, and I don't believe I am (not my personality, my code or my lifestyle). It is also very easy for you to get a dedicated audience with Chinese users for rants like this, and I doubt you will take my opinion written here back to your baidu thread, but do consider it.
While your code may not be perfect, I'd like to thank you for your contributions to osu! should you decide not to continue helping with development. From my end, while we didn't agree on everything, I believe that I tried as hard as I could to help you out (buying dev hardware, offering a more full-time position if you ever wanted it) and did give you complete freedom as to what you worked on, even though I didn't always merge your stuff.
For those looking on at this, note that I wouldn't usually post a response like this publicly, but the original was in the public domain so I guess this is the best way to handle it.
Background: I had to remove woc from the dev team a few days ago after I was alerted by a large number of emails that he was deleting all his code from the osu! repository. I did this as a safety measure and still have plans on letting him return should his mind change.
woc posted a long rant on baidu (http://tieba.baidu.com/p/2905827484?pn=1) to which I will provide responses to. Yes, I had to use google translate to retrieve meaning, but I believe that for the purposes at hand that should be fine. I am avoiding any of the emotional language as that is not relevant.
update: better translation in this post. I won't update my responses below as I don't believe it changes too much, but does allow for a much smoother read for those interested.
Today you can see osu core functionality of the code, 70% is in the first month of the first to write almost no change, including the editor, song selection, play relevant and so on. Since ppy was caused by mistakes on the young architecture has been left inside.This is generally how software works. If you have been following my commits, I have been striving to fix the problem of archaic code by improving quality of existing code.
What is it about the mistakes, editor of the core code, I never did not understand too, ppy said he is still able to understand, but he just can understand, change does not move. So encountered want to change the function editor, I always have to spend a lot of effort, carefully avoiding the possible impact on the normal use of the code, write new logic. More often, simply can not add things.Yep, we did discuss this. And you've done a great job working with my shitty code.
If you are a player, so the editor used to cool unhappy okay with you, then you definitely related to the stability of the game can not stay out of a game program on 60fps also felt card, which in itself is a problem, only 60Hz monitor refresh rate . To solve this problem, we need to plot the underlying logic and game logic, with two threads to run. This is the only possible reconstruction osu completed.I can't figure if you mean the drop to 30fps when the game isn't focus (intended) or just performance in general. There are always methods of increasing performance, but your suggestion of moving updates and draw calls to separate threads – especially in a game like this – will introduce a hell of a lot more issues (in the form of very hard to track bugs) than it is worth. I'm not sure if you've ever tried dealing with a complex multi-threaded code-base, but it ain't easy.
Besides dev team is how it works with the code to compile and release the power of one person only ppy, I wrote the code, he should look at every sentence. Prior to September 13, my code is in an independent branch, all I can quickly change the bug into submission, then ppy periodically merge once, then release. Prior to this, we work very well, but then, I have to follow the requirements of ppy, every alteration, to create a new branch, separate submission.This was the requirement to use pull requests. Even after then, I haven't looked through every sentence as I hoped to (I believe ALL code should be checked by a second set of eyes in all cases), but I have been able to hold back some of your patches where the quality wasn't really up to my standard. Also you mention the mania HR mod, but I was pretty sure the last thing you told me was to not merge it because it isn't ready. You probably should have brought this up again...
ppy has a habit is something of interest only, and will start a very good idea, and use it as motivation to do it without food or drink, and then osu is doing it, so that also the case modding v2 . The result is almost unanimously opposed BAT, because although BAT when nothing good reward, but at least do not waste their time with ppy justice guards play this game.I don't believe anyone is opposed to v2. And yes, it is taking longer than expected but it will happen.
The results did not compromise the final ppy, modding v2 did not die, but ppy greater stir, he wanted to reconstruct the official website (ppy so it may be able to osu compared to code, can not endure the official website, the official website of the code does not tucao more) and then go together to make modding v2. The official website is currently ppy spend money on hiring people to write, slow progress.
So you see a shield SB later ppy function, in fact, that is lying to you, so that you shield on the surface, in fact, the shield SB ppy and background darkens when treated like cheating, documented. His bones are not recognized by this behavior, although it regardless, but maybe someday pp increase in the algorithm in addition to buckle off the SB and the like.Storyboard disabling was being logged (maybe you refer to this as "being treated like cheating") so I could get an idea of how many people were disabling/dimming things. I used this data to build up statistics and come to the final conclusion of allowing it in all cases. I think you may have misunderstood my intentions here.
Speaking pp, will have to mention ppy players really do not know, one of these two years ppy do three big (stupid) thing is replaced by pp score, As this technology flow algorithm itself is good, tom We have to prove that, but nothing to do with ppy.I hope I did not take credit for the new pp changes anywhere. And yes, I admit that there is no way I can keep up with difficulty/scoring/pp algorithms myself, because it is a very time-consuming tasks to understand the whole picture. This is why Tom has taken over now.
It score event, I realized ppy just a developer, with the players not detached. So no matter how he was doing, related to the average player does not matter, he was just doing what he likes.I feel I do my best to keep up with player's demands. I am sorry if you don't see it that way.
osu in the plug-in ppy eyes is two, one is able to detect, one is not detected. Can be detected, and will ban, but never once ppy ban, according to his logic, immediately ban hanging with plug to get into the war, he did not have the energy to play cat and mouse, so the delay ban hanging, probably open people are afk then hang it.With regards to your issues with the banning system, I really have nothing. Yes, in many cases we delay bans until we have enough information to deal with them, and do this in a way the team can handle the huge workload (which sometimes means people aren't banned as soon as they otherwise could be). My aim isn't to ban all cheaters, but to make it so the average player is not affected by cheaters. This is already hugely time-consuming and I can't really offer more.
Finally had to mention cookies.I still have no idea about Cookiezi. He caused shit and used cheats, therefore is banned. He never tried to contact me, so I can't really say anything on this matter (maybe you know more than I do).
It was written a long one in loc posted to ppy to see, then ppy banned mania without sound. ppy think is my indulgence, and then very harsh to say: You do not want to engage in sound mania to roll out their own to do a Drop simulator, osu is sound, the sound is osu, which I absolutely do not compromise.I'm still not willing to budge on the hitsounding issue.. bemani* games are the most IMPORTANT ones to properly hitsound.
The second sound contradictory or mania, but this problem is not, as is ppy suddenly some of my key code comments, and then put someone in the forum did not sound feedback mania bug stickers tossed me, let me go and change.Woah woah woah. Let's examine what your code changes actually did: you were loading EVERY mp3/wav in the song directory every time the beatmap started. This included the mp3. The song mp3, not a sample. I honestly don't like calling people out so avoided doing this, but this was adding between 2-20 seconds load time every time a map is played. The song was being loaded into memory and stayed there until play was finished.
I accepted that this was my fault for accepting your code changes which caused this.
Is followed two days later, ppy start deleting my other code, this time there is no reason to delete, and then my last as a shake M, say one thing: Please tell me the reason to remove it.I deleted one huge method of yours from the code in my maintenance branch, yes. It was a copy-paste of the most VITAL code in all of osu! with almost no alterations. You made a second copy of the whole .osu loading code just to change a few minor areas and allow it to be loaded into the editor in a different way.
I was working with that code for an hour or so before realizing I was editing your method rather than the one actually being used. This is very hard to see due to it containing the same 500+ lines. Therefore, I removed your method to avoid future confusion. Tom has already implemented what you were doing using a much nicer solution (as he required similar behaviour for difficulty calculation).
No answer.I didn't answer you because I was struggling internally. I know that you don't like it when I call your code bad, but I had no other way to explain this particular case. I am sorry for not being more verbose, but I thought the outcome would be better if I just went ahead and fixed it on my branch. I think you may have taken that the wrong way.
Also you fail to mention the number of times you submitted patches which wouldn't compile (you never tested that it compiled before submitting). Or the recent bug fixes you have submitted which almost always cause a really obvious regression that can only mean they were never tested.
Overall, players choose osu, really because there is no other game on the pc platform there are so many songs, the sound itself is a very small minority tour player, rhythm masters players millions of scale, it is almost full China is willing to play all audio and more than 50% of the number of trips. Number of core audio tour playing even 1% of this base not to.While I am grateful for how popular osu! is, keep in mind I am not at all worried about its future. I do not care for market share nor profit. I only care about making osu! as enjoyable for users as I can. If you don't feel that this is the direction osu! is heading, and don't wish on helping further that cause, then I strongly suggest that you start from scratch.
Yes, I can be stubborn when I am confident that my opinion is right, but this is how I choose to run osu!, and it lets me run it in an efficient way.
I'd just like to end this by saying that it is *very* easy to call a person out like this. No one is perfect, and I don't believe I am (not my personality, my code or my lifestyle). It is also very easy for you to get a dedicated audience with Chinese users for rants like this, and I doubt you will take my opinion written here back to your baidu thread, but do consider it.
While your code may not be perfect, I'd like to thank you for your contributions to osu! should you decide not to continue helping with development. From my end, while we didn't agree on everything, I believe that I tried as hard as I could to help you out (buying dev hardware, offering a more full-time position if you ever wanted it) and did give you complete freedom as to what you worked on, even though I didn't always merge your stuff.
For those looking on at this, note that I wouldn't usually post a response like this publicly, but the original was in the public domain so I guess this is the best way to handle it.
