I did think that many of the games onubense vs APU_1 were much better than maybe should be for player I beat 10.5 - 2.5, but not I was sure, so not I did say anything. Onubense did play some bad line in first half of match even if APU_1 was computer, so was hard of to say of 100%. Still, not this does surprise me as much as would maybe I like because of how strange it is for player I have beat much to have such good luck of with Onubense. Not I am actual sure whether I would like of to play computer or Onubense more if I win next round because both are very difficult in different ways.
"Its impossible that i have used computer assistance when i dont have used it. Even so, i have seen my games and its obvious that i didnt use it in any game, with mistakes, blunders and big mistakes." - APU_1
Then maybe we would be okay of to examine these games of you with Onubense? I will go through and find actual number of "real" mistake and blunder that are appear since sometimes lichess is not like opening or is not like move even though both is are good if analyze longer. Same is for choose simple endgame win over fast win that are complicated. Also will try find anything that not look such human.
1. lichess.org/aVpKwu4F No real inaccuracy if not count opening, and 14. Rc1 not is actual mistake and is choice 2 of stockfish, although also is worth of say that 6. b4 is best move of accord to stockfish. Not it is most natural move though. 8. Nc3 also is mistake, but is both natural move and is move that not does stockfish see is mistake at first, so will say is only inaccuracy. Is difficult for human of to see why is mistake with 8... h5! though. 9. b3 is inaccuracy if are ask me, since is only 0.5 pawn worse than 9. d4, but still is mistake. This is only real problem lichess analysis find. Also note that APU_1 was spend time on first moves 4. c3 and 5. cxd4. Is seem strange that not white would know moves of own opening in main line for black, since would want play opening as fast as can in important game like this. 5... d5 is a little strange, so can understand spend time on 6. Qh5 though. Also, not is a lot of time you spend here on these two moves. At end, I think is actual 0 Blunder, 1 Mistake, 2 Inaccuracy. Is suspicious, but not is have flat move times, and could see this be played of by good player.
2. lichess.org/NBpJ7q1W Okay, this is interesting game. Often these games of this lines are very complication, and are harder of black even though maybe not it should be that bad if are play perfect. Also, normal is not of to play these line and to instead play of either 2... d5 or 2... Na6. 2... e6 very strange of for human to choose, but maybe I think APU_1 could choose because know these lines well and is complication. Also is worth to say that at some depth 2... e6 is top move of stockfish at certain depth even though it realize that is better of play other move if analyze more time. Also moves at beginning were not such much time is used as was in game 1. First thing of real suspicious I find is maybe 6... g5. At first is move stockfish like of most, but later not it is. Is also look like very dangerous move as human. Much more dangerous of 6... g6 which are actual better with many of same idea. Maybe could be misclick though. It at least are inaccuracy even though is stockfish inaccuracy also. Also appear that are inaccuracy of 8... Nd7, even though again is inaccuracy that stockfish suggest. Also it are very normal move of for a human. From here though, is very strange. Not only does play very well, but play all top stockfish moves until move 21... Ba6+ which is clear win for any player. Not many of the move were super hard of to find, but is very good play for all stockfish top move choice every move until end in such position when are can be tricky of find correct ideas. So in total is only 2 inaccuracy in position with many idea, both of these inaccuracy also were top stockfish move when analyze not much time. Very suspicion, but not is impossible, since many of move not were such hard of to see.
3. lichess.org/7BJB3Szd Again, is throw off by 6... Nf6 and 7... d6. Would think that would at least know some of lines that plan of to play vs onubense in important match for some slightly not mainline ideas... Still, maybe not did think that onubense not would play main ideas, so not did think would need know such lines? Also of the twelve inaccuracy that are listed, only three are not in clear won part of endgame or in opening. Of these are we have 15. Nb5 which are move that try to trade and still is winning, 16. Nxc7 of with same idea, and 22. exd5. Again, still winning and with idea be trying to set trade piece for more simple win. Still, not they are best moves, so maybe have 3 inaccuracy, 0 mistake, 0 blunder. Are a couple tricky move to see that are in rook ending, but not rest is too suspicion. Not too many strange moves of for human to see or for too many of strange idea. Are some, yes, but other than find a few good move and not blunder not is such strong evidence of cheat. So is strange, but again, not would I think is mean for sure is cheater.
4. lichess.org/2lYaKT2p This is first game that APU_1 are lose, and make strong case that him is are human, at least of for this game. Are many mistake and all of moves are normal idea of for black. Not too much of time was spend on opening, is clear that was lost when still have had time of on clock. Is look very human with many mistake and blunder of for both player.
5. lichess.org/snvk0qfu Not is play such slow of opening that should know, but is also because of maybe play of last games and remember? Is of interest that outside of choose of opening, only inaccuracy is again of 14. b3 ideas, so maybe is because knows this is idea of some other position? Is also worth of say that all of moves after 5. cxd4 until 13. Kh1 are best of say stockfish. 13. Kh1 is second best move of say stockfish, but is very strange move of for human. Not I do see idea of immediately, but maybe someone else is are see idea of this move. Maybe are to try play e4 with not have king on g1-a7 line? Not I am sure. 14. b3 at least it are clear of what idea is, but 13. Kh1 look of like very strange move of human to me. After this, are many top engine move again, but not they are appear such impossible of to find on own even though all together are difficult and stockfish top move stop at 22. Rc1. Some of moves are hard to understand why take such much time if did not see plan before are play, but can understand if were try be careful. Again, suspicion but not I am sure this are why this are such suspicion other than find play well and not have clear errors other than have kept up for such far as this without any clear error other than one game where he are lose.
Maybe am will see if something is more obvious later when do last 5, but so far if look at games 1, 2, 3, 5 is very suspicion, but not I can see clear reason for to say cheat other than that all four are not have any real error and that is play extremely well against other player who are should be very, very difficult of for him to win of game with. Am sure that similar case could of be made with for me if choose some match I play against other player. However, is strange that would happen here since not this is go through all matches to look for one where he play best, but is just look at one match here, and are much more unlikely to happen of sample of one match instead of many match.
"Its impossible that i have used computer assistance when i dont have used it. Even so, i have seen my games and its obvious that i didnt use it in any game, with mistakes, blunders and big mistakes." - APU_1
Then maybe we would be okay of to examine these games of you with Onubense? I will go through and find actual number of "real" mistake and blunder that are appear since sometimes lichess is not like opening or is not like move even though both is are good if analyze longer. Same is for choose simple endgame win over fast win that are complicated. Also will try find anything that not look such human.
1. lichess.org/aVpKwu4F No real inaccuracy if not count opening, and 14. Rc1 not is actual mistake and is choice 2 of stockfish, although also is worth of say that 6. b4 is best move of accord to stockfish. Not it is most natural move though. 8. Nc3 also is mistake, but is both natural move and is move that not does stockfish see is mistake at first, so will say is only inaccuracy. Is difficult for human of to see why is mistake with 8... h5! though. 9. b3 is inaccuracy if are ask me, since is only 0.5 pawn worse than 9. d4, but still is mistake. This is only real problem lichess analysis find. Also note that APU_1 was spend time on first moves 4. c3 and 5. cxd4. Is seem strange that not white would know moves of own opening in main line for black, since would want play opening as fast as can in important game like this. 5... d5 is a little strange, so can understand spend time on 6. Qh5 though. Also, not is a lot of time you spend here on these two moves. At end, I think is actual 0 Blunder, 1 Mistake, 2 Inaccuracy. Is suspicious, but not is have flat move times, and could see this be played of by good player.
2. lichess.org/NBpJ7q1W Okay, this is interesting game. Often these games of this lines are very complication, and are harder of black even though maybe not it should be that bad if are play perfect. Also, normal is not of to play these line and to instead play of either 2... d5 or 2... Na6. 2... e6 very strange of for human to choose, but maybe I think APU_1 could choose because know these lines well and is complication. Also is worth to say that at some depth 2... e6 is top move of stockfish at certain depth even though it realize that is better of play other move if analyze more time. Also moves at beginning were not such much time is used as was in game 1. First thing of real suspicious I find is maybe 6... g5. At first is move stockfish like of most, but later not it is. Is also look like very dangerous move as human. Much more dangerous of 6... g6 which are actual better with many of same idea. Maybe could be misclick though. It at least are inaccuracy even though is stockfish inaccuracy also. Also appear that are inaccuracy of 8... Nd7, even though again is inaccuracy that stockfish suggest. Also it are very normal move of for a human. From here though, is very strange. Not only does play very well, but play all top stockfish moves until move 21... Ba6+ which is clear win for any player. Not many of the move were super hard of to find, but is very good play for all stockfish top move choice every move until end in such position when are can be tricky of find correct ideas. So in total is only 2 inaccuracy in position with many idea, both of these inaccuracy also were top stockfish move when analyze not much time. Very suspicion, but not is impossible, since many of move not were such hard of to see.
3. lichess.org/7BJB3Szd Again, is throw off by 6... Nf6 and 7... d6. Would think that would at least know some of lines that plan of to play vs onubense in important match for some slightly not mainline ideas... Still, maybe not did think that onubense not would play main ideas, so not did think would need know such lines? Also of the twelve inaccuracy that are listed, only three are not in clear won part of endgame or in opening. Of these are we have 15. Nb5 which are move that try to trade and still is winning, 16. Nxc7 of with same idea, and 22. exd5. Again, still winning and with idea be trying to set trade piece for more simple win. Still, not they are best moves, so maybe have 3 inaccuracy, 0 mistake, 0 blunder. Are a couple tricky move to see that are in rook ending, but not rest is too suspicion. Not too many strange moves of for human to see or for too many of strange idea. Are some, yes, but other than find a few good move and not blunder not is such strong evidence of cheat. So is strange, but again, not would I think is mean for sure is cheater.
4. lichess.org/2lYaKT2p This is first game that APU_1 are lose, and make strong case that him is are human, at least of for this game. Are many mistake and all of moves are normal idea of for black. Not too much of time was spend on opening, is clear that was lost when still have had time of on clock. Is look very human with many mistake and blunder of for both player.
5. lichess.org/snvk0qfu Not is play such slow of opening that should know, but is also because of maybe play of last games and remember? Is of interest that outside of choose of opening, only inaccuracy is again of 14. b3 ideas, so maybe is because knows this is idea of some other position? Is also worth of say that all of moves after 5. cxd4 until 13. Kh1 are best of say stockfish. 13. Kh1 is second best move of say stockfish, but is very strange move of for human. Not I do see idea of immediately, but maybe someone else is are see idea of this move. Maybe are to try play e4 with not have king on g1-a7 line? Not I am sure. 14. b3 at least it are clear of what idea is, but 13. Kh1 look of like very strange move of human to me. After this, are many top engine move again, but not they are appear such impossible of to find on own even though all together are difficult and stockfish top move stop at 22. Rc1. Some of moves are hard to understand why take such much time if did not see plan before are play, but can understand if were try be careful. Again, suspicion but not I am sure this are why this are such suspicion other than find play well and not have clear errors other than have kept up for such far as this without any clear error other than one game where he are lose.
Maybe am will see if something is more obvious later when do last 5, but so far if look at games 1, 2, 3, 5 is very suspicion, but not I can see clear reason for to say cheat other than that all four are not have any real error and that is play extremely well against other player who are should be very, very difficult of for him to win of game with. Am sure that similar case could of be made with for me if choose some match I play against other player. However, is strange that would happen here since not this is go through all matches to look for one where he play best, but is just look at one match here, and are much more unlikely to happen of sample of one match instead of many match.