SparkysAdventure Posted June 5, 2021 Posted June 5, 2021 Impressive clocks! Do GDM Off, tRAS 21, and tRC 29 all not work for you? You may want to try higher fclk or unlinked. Is this a bugged run? Quote
GtiJason Posted June 5, 2021 Posted June 5, 2021 2 hours ago, Sparky's__Adventure said: Impressive clocks! Do GDM Off, tRAS 21, and tRC 29 all not work for you? You may want to try higher fclk or unlinked. Is this a bugged run? ? ? Memory Latency 42079 10.3 ns 1 Quote
Dr.Antoine Posted June 5, 2021 Posted June 5, 2021 Hi it is not a bugged run I just found goods parameters to have the best results, no I optain the best performance with theses settings it is because I give it to the community 1 Quote
Crew Leeghoofd Posted June 5, 2021 Crew Posted June 5, 2021 one does not jump from 11K to 18K in memory score without bugging the benchmark... Provide the maxmem setting plz ? I hope you don't believe that your parameters allow you to beat scores with the same hardware clocked 500-700Mhz higher... 1 Quote
Dr.Antoine Posted June 5, 2021 Posted June 5, 2021 Maxmen 1G, if you want I can give you my tips in private 1 Quote
Crew Leeghoofd Posted June 5, 2021 Crew Posted June 5, 2021 That's where it comes from... at G.Skill booth we did not accept below 1.5G to prevent bugged outputs.... 1 Quote
Dr.Antoine Posted June 5, 2021 Posted June 5, 2021 It is just the ryzen 5000 architecture, there is only 2 thinks to optain this low latency, find the good core and limitate the amount of ram 1 Quote
Crew Leeghoofd Posted June 5, 2021 Crew Posted June 5, 2021 it also works on Intel... geek bugs with too low available memory, seems I will need to go and clean up some scores... 1 Quote
Dr.Antoine Posted June 5, 2021 Posted June 5, 2021 it is just the optimisation of this benchmark not a bug, you can try the same think : 1G ram, desable SMT and put the affinity on the core 2 ... Quote
Crew Leeghoofd Posted June 5, 2021 Crew Posted June 5, 2021 I fail to see the optimisation of the benchmark when you boost 80% in memory score and you beat way higher clocked setups. It's intentional bugging the output... Quote
Dr.Antoine Posted June 5, 2021 Posted June 5, 2021 so I can do a score with 2G maxmem it is good ? I think it could be mentioned in the rules Quote
Crew Leeghoofd Posted June 5, 2021 Crew Posted June 5, 2021 Well perform a test, change maxmem to 1.5GB & re-run, afterwards set 2G maxmem & re-run. For me this is common sense something to grasp something is not right. Either the Top 5 in that ranking suck monkey balls or you are too darn good... 1 Quote
Dr.Antoine Posted June 5, 2021 Posted June 5, 2021 ok I m going to do a 2G and a 1.5G score with the 5600x and the 5800x can you remove my scores GB4 single of this two processors please Quote
Dr.Antoine Posted June 5, 2021 Posted June 5, 2021 and titane used the same tweak for the 5800x on GB4 Quote
Dr.Antoine Posted June 5, 2021 Posted June 5, 2021 but I look to many score with theses tweaks so I think now it is a commom trik Quote
cbjaust Posted June 6, 2021 Posted June 6, 2021 13 hours ago, Leeghoofd said: it also works on Intel... geek bugs with too low available memory, seems I will need to go and clean up some scores... GB4 Single Core scores with <1.5GB Maxmem as far as the current global top 30: https://hwbot.org/submission/4753151_rauf_geekbench4___single_core_core_i9_11900kf_11450_points/ https://hwbot.org/submission/4726111_safedisk_geekbench4___single_core_core_i9_11900k_11409_points/ https://hwbot.org/submission/4752362_sergmann_geekbench4___single_core_core_i9_11900k_11363_points/ https://hwbot.org/submission/4722432_noizemaker_geekbench4___single_core_core_i9_11900kf_11232_points/ https://hwbot.org/submission/4511897_davestarrr_geekbench4___single_core_core_i9_10900k_9919_points/ https://hwbot.org/submission/4142526_ale_belo_geekbench4___single_core_core_i7_8700k_9868_points/ https://hwbot.org/submission/4194819_xmec_geekbench4___single_core_core_i7_8700k_9780_points/ 1 Quote
Guest endpunkt Posted June 6, 2021 Posted June 6, 2021 Does not always show 1000mb either. Tested just now. Quote
Crew Leeghoofd Posted June 6, 2021 Crew Posted June 6, 2021 Most logic step might be to remove globals sadly... 2 Quote
Guest endpunkt Posted June 6, 2021 Posted June 6, 2021 This is the third biggest safedisk. Looks fine Quote
Guest endpunkt Posted June 6, 2021 Posted June 6, 2021 6 minutes ago, Leeghoofd said: Most logic step might be to remove globals sadly... Reason we have Mat and benchmate. Loop hole wizzards are gonna loop hole estimate 25% of top scores 50. 1000mb done. Quote
Guest endpunkt Posted June 6, 2021 Posted June 6, 2021 I'm down to 10K scores. You have x4 naughty boys so far ? Quote
SparkysAdventure Posted June 6, 2021 Author Posted June 6, 2021 To my knowledge, this is a consistent tweak, I think relevant for Ryzen. 2 1 Quote
Crew Gunzi Posted June 6, 2021 Crew Posted June 6, 2021 (edited) 7 hours ago, endpunkt said: This is the third biggest safedisk. Looks fine Reported this exact score back in November due to the insanely low latency, was told it was fine and a tweak Keep said he wanted to reproduce score sometime but did not want to share the tweak Edited June 6, 2021 by Gunzi 2 Quote
Guest endpunkt Posted June 6, 2021 Posted June 6, 2021 2 hours ago, Gunzi said: Reported this exact score back in November due to the insanely low latency, was told it was fine and a tweak Keep said he wanted to reproduce score sometime but did not want to share the tweak At this stage that one clears. It's look for obvious ones first 1000mb done _____________ https://hwbot.org/submission/4731907_kansaï_geekbench4___single_core_core_i9_11900k_9358_points/ When I was running my processor. I'm looking at this score, mems are at 4900 Does not mean it is a bad score like some of the others. It's extremely fast + fast timing. BUT. Done 1000mb. Even 2000 max mem it may not run. That is the point, not equal to others doing bench correct Quote
Recommended Posts
Join the conversation
You can post now and register later. If you have an account, sign in now to post with your account.