This Software KILLED My FPS! - Process Lasso Gaming Performance Tested
Vložit
- čas přidán 2. 06. 2024
- Check out this comprehensive video where I put Process Lasso, a popular software for optimizing gaming performance, to the test. Many users claim that it enhances gaming performance, particularly on processors with hybrid architectures or small e-cores. This is a software which I’ve been hearing about a lot since the early Zen 2 days when AMD brought chiplets to the market. Now that both manufacturers are using hybrid architectures for the mainstream desktop segment (ie: 3D VCache, E-cores) the use of this program is becoming more prevalent amongst gamers. In this video, I benchmarked 12 modern games using Process Lasso to ensure they ran exclusively on the P-cores, bypassing the e-cores. The results were surprising. I had tested e-cores and hyper threading disabled through the bios, so I had an idea of what to expect but was still quite shocked to see these results. Find out the detailed performance analysis and my recommendations for using Process Lasso in the video. Test things for yourself and make an informed decision for your system configuration. Stay curious and enjoy your gaming experience!
TESTS Specs:
CPU: Intel Core i9-13900K 5.7GHz Pcores, 4.6GHz Ecores, 5.0GHz RING
CPU Cooler: MSI MAG Coreliquid E360
RAM: Teamgroup T-creat 32GB(2x16GB) 7200CL34 tuned timings
Motherboard: MSI Z790 Carbon Wifi
SSD: Corsair MP600 PRO 4TB
GPU: MSI RTX 4090 Gaming X Trio 3GHz cores, 24gbps mem
PSU: EVGA 1000G3
Check out my 13900K Hyperthreading 40 Game Benchmark Vid
• Is Hyper-Threading Use...
Watch my 13700K E Core ON vs OFF 40 Game Benchmark video
• I Did Not Expect These...
Watch my RTX 4090 vs 3090 42 Game Benchmark video
• RTX 4090 vs RTX 3090 M...
Watch my RTX 4090 at 2GHz video
• If Nvidia Didn't Incre...
Buy an RX 7800XT Here:
amzn.to/3EuQYoh
Support the channel by donating
**** www.paypal.com/donate/?hosted... ****
Want to build a Gaming PC? Use my Amazon Affiliate link below!
United States - amzn.to/2JKq8uW
Canada - amzn.to/2GYSkbM
United Kingdom - amzn.to/2qupqda
Follow me on Twitter
/ dannyzreviews
Timestamps:
00:00 Intro
01:37 Scheduling Issues
02:34 Why Process Lasso
04:27 How it works
06:36 Test Specs
07:03 Benchmarks
10:55 12 Game Average & Conclusion
#PCgaming #optimizegamingpc #ProcessLasso #boostfps - Věda a technologie
The reason why process lasso doesn’t work with Intels architecture is the thread director that handles it underneath. That and process lasso is manual rather than a soft affinity. Process lasso works great for AMD chips cause the thread director doesn’t get in the way. It basically confuses the scheduling and latency. APO is basically process lasso from the way of what we are trying to do. But with intel you can’t tune it manually. You have to buy an entire new Chip and PAINFULLY wait for them to release apps for it.
I also do love the “verify for yourself” testing 100% on my channel as well too cause sometimes weird bugs and things get in the way that you may not be able to explain. Good video!
Windows has something called CPU sets which is a more soft affinity, maybe that works better, unfortunatly that's only supported by the pro version of process lasso
What you are missing about Process Lasso is what most of us use it for. Instead of trying to modify your games, you should be using it to prevent other background processes from "running wild." This is where the program shines. It was never designed to be the "game fixer" it was designed to prevent processes from going crazy in the background. Using it this way, I see processes that it has paused, it doesn't cause any issues in my gaming (I game a lot), and everything stays smooth and responsive at all times.
Not a single non virus process should ever start "running wild" in the "background".
And if you really need to restrict a process, use the windows eco mode or just limit it to a single thread.
(both of which have been part of the task manager for ages)
@@Waldherz If you claim to actually use Windows and say that processes don't run wild, you are trolling. Those of us that have been here since Windows was something we started from the command prompt in MS-Dos know that there have always been issues with this. Programs leak memory, they crash and take the system down, they randomly decide to do what they want when they want, etc. I am more than capable of managing Windows without Process Lasso, but I find it to be convenient in what it does.
My computers are built in house to my specifications, and I can count the number of blue screens I've had in the last couple of years on 1 hand. So I would ask you not to come at me like I don't know how to manage my systems. Computers are tools, as are the programs we run on them, and when we find a tool that makes the job easier so be it.
A wall of text or claimed age doesnt make you right.@@Belanthiel
@@Waldherz So if you have nothing to say, or anything constructive to back yourself up I assume our interaction is closed!
@@Waldherz yeah, claimed age does not make someone right, but... My Glob, gen Z is afraid to read. 7 lines is not a wall of text not here or in Liliput, where Gulliver was tied to the floor by little people
great fucking video, I've been seeing people mention process lasso for a while now and this is exactly the data I was looking for. Thanks man, keep up the good work, you just earned yourself a new subscriber.
What if you use CPU sets instead of affinity (a windows feature which is basically a more "soft" affinity) unfortonatly it's only supported by proccess lasso pro as far as I know? Also what happens if you also set every other process to only run on e cores?
Great video like always danny! bravo
Thank you for the benchmarks and the explanation of everything, recently Quick CPU have got an update which introduces an heterogenous policy for processors with hybrid architecture like Intel 12 th gen and so on, I wanted to ask if you can try it and if it adds stability. Keep up the good work hope your videos reach majority of people
Process Lasso still has many use cases for intel processors. For example you could set Discord, steam and other background programs to run only on E cores which can minimize stutters. Also disabling "Pro Balance" might be helpful.
exactly!!! this helped my gaming so much
Have you tried Process Lasso with pro balance enabled but without disabling e-cores? It smoothed out the frametimes and eliminate some stutter I had in Hunt Showdown.
can u give me tutorial?
Did you use Probalance when using process lasso? ive noticed framedrops and stuttering when using that setting without it everything seems fine.
Same, I disabled this setting, and instead of affinitnity process, I used cpu sets because, for some reason, halo infinites chat won't work with affinitnity
Really surprising tbh.
Thanks for the video.
May I ask you to test CS2 please ? As source 2 engine is using only 4 cores, so could be interesting to see.
Hey Danny Awesome Video Man! I still watch all of your content, and really enjoyed this one. I wanted to ask, in case anyone else doesn't, if you would like to do the same test methodology but on a Ryzen 7000 Based System? I used to use Process Lasso back in 2000 and 3000 series Ryzen Days, but stopped using it for 5000 and 7000 series, and am interested to see if you observe a benefit from using Process Lasso on a newer dual Chiplet based CPU from AMD.
Anyways, thanks for the awesome content as always!
Very nice video, thank you! I have an asus rog strix g16 with an i9-13980hx, all the testing was done in time spy due to being the same and having to guess if the change was due to different locations inside the game, so basically in my experience as well was that it worked best only with soft rules like CPU Sets rather that hard rules (cpu affinity), in that I managed the best score. That indeed is different results than of the people making videos about it, I imagined it was probably due to the new architecture...
SO I'm a bit confused, when you say "stock, no lasso" are you comparing Lasso on forcing P-core only, and also turning off E-cores in bios, or leaving the e-cores enabled?
Would have been nice to see a comparison of PL forcing P-cores, vs APO (I know, 14 gen only and limited games), vs Bios E-cores OFF, vs "doing nothing" (stock, with E-cores ON, APO disabled).
Are you using any of the other features from Process Lasso or Bitsum, ie the power management stuff?
What about older games that are SUPER cpu bound? One I can think of is Digital Combat Simulator, that game eats CPU's alive, especially in VR.
Thank you for your efforts in actually testing the software. To reveal there was no uplift using the software vs without is immensely helpful.
Rather than post how this software so amazing, without actual benchmarks, I would welcome anyone else to do the same and see if their findings are similar. I have used PL for a short time and can say that there was less issues in terms of frame time and stuttering. However, the one reason why I stopped using PL is because it lowered my PCore turbo clocks, and matched it to the ECore clocks. I would reboot my system, and the turbo clocks would load in, but as soon as PL launches, it removes the turbo clocks. I may or may not have tweaked some setting that prevents this, but that is behavior I didn't expect.
Once I uninstalled the software, the turbo clocks came back.
This video is really helpful, I got it recently because Elden ring stutters like crazy on newer intel CPUs with e-cores enabled and was fixed with Process Lasso, I could go with the turn e-cores option but its easier to have project lasso not run until opened and only open it when games have issues then rebooting to turn e-cores off. but thanks to this video I'll probably leave it off till such use cases.
I use Process Lasso for various reasons besides gaming, but recently I ran a benchmark on Guardians of the Galaxy to see if it did anything that was consistently noticeable in results, and it did.
I don't do any of the E-core adjusting, the only option I turn on in Process Lasso is the Performance Mode/Bitsum Highest Performance Mode, and that does make a difference in CPU utilization. My FPS was essentially the same, but my CPU Max rendering time in (ms) was 36.1ms without Process Lasso being used, and then 20.2ms with it on.
I want to do another test when I get a chance and use a more detailed stats showing so as to see what exactly is going on. It's possible frametimes may be better overall with it on, or at least better in certain circumstances/games.
CPU is a i7-13700k by the way.
how do you check CPU max rendering time ?
Try using CPU reserved sets (tool made by Amit) to isolate drivers/tasks from CPU cores then using process lasso. It works well. Saves a lot of time from assigning affinities to every single process
Doesn’t work with games that use Easy Anti-Cheat because you cannot reschedule the games onto p-cores.
Hello please make some test without cpu 0 in games, and process lasso have 2 options: PROBALANCE and PERFORMANCE MODE, can you tell us more about these options ?
I use device manager. I have an office pc with a 13700k, 3090 & gtx 1050ti. I like how the Intel cpu (13700k) multitasks. I mine with a 3090, use a gtx 1050ti for video/playback. I cpu mine with the cpu's P-cores and play Batman: Arkham Asylum (older games) at 1440p all high settings locked at 62fps using the cpu's E-cores and Intel igpu (overclocked to 2000mhz, stock voltage) while playing/listening to a CZcams 1080p video, all with no stutters.
Please redo this test with a newer generation AMD cpu as I’d love to see if Lasso is AMD architecture biased (works well) or there’s an issue with Intel (works worse as per this video)CPUs as Microsoft’s Windows scheduler is known to have built in intel P/E-core support.
I'm on a 7600, and empirically I felt a slight-but-noticeable increase in input delay with Lasso. I play fighting games at a pretty high level, and I would notice myself messing up muscle-memory timings by doing the inputs too late. Force-disabling Lasso stopped that weird feeling.
Realistically, it's probably dependent on the application though.
@@darkassasin111 youre quite right about it being program dependent. I managed to get some more frames and faster loading times by telling process lasso to use core number 5 for the notoriously bad gta 4 pc port which doesn't really do multithreading. I'm on a 7600x and i only use Plasso for increasing single core performance and remembering program affinities.
@@pastyboi2789 Yes it does seem to help on older single threaded games but with modern processors you can just muscle your way through because single core performance is way above what it was 13 years ago especially when compared to AMD's Bulldozer architecture which had truly abysmal single core performance
It's made a nice difference in snappiness on my 5900x CPU.
Plus SOME games prefer to run on CCD0 only, like the latest Forza Motorsports. I can a few fps by limiting it to just CCD0 (6c/12t) instead of all 12c/24t. I also do some tweaking and move a lot of apps onto cores 10 and 11, the lowest performing ones, and have that set in a gaming profile. So, that's 2c/4t for Steam, EA App, etc. to all run on instead of using Core 0.
Sure, the CPU is powerful enough and my GPU isn't the greatest (a 6700XT 12GB) to notice 20fps+ boosts, but I do get boosts in the right games.
And with Project Lasso running, just letting it do its normal optimizing, everything truly is FAR snappier. Opening browsers, going through Radeon Software.. just all the little things like that.
I like the app. It gives me control over my CPU, which I can do with task manager, but it saves the settings so I don't have to redo them every time. Plus I can make profiles which is great. One for normal usage (browsing the web/watching CZcams/etc.) vs gaming.
I use it to force all my games onto one CCX with my 5950x and I get more consistent framerates, not necessarily a huge FPS increase.
Great video. I'd love to see a similar video for the AMD 5800X3D. I'll stay away from Lasso until then. Thanks for sharing your findings, even tho they may not apply to my scenario.
I would really like to see someone compare 7800x3d to the 7950x3d with process lasso as well!
Very insteresting video. The main reason why I'm interested in this program is because older games like Wolfenstein 2009, and Far Cry 3 and 4 either have horid performance or wont launch at all on processors with more than 4 cores. Far Cry 3 and 4 just hover around the 70 fps range on my rtx 3080 and Ryzen 5950x and Wolfenstein 2009 just wont launch at all unless I do an obscure command line argument that apparently deals with CPU affinity. Changing the CPU affinity also helps with games like The Witcher 2 apparently. I would love a video testing out old games that have problems on new hardware with this program.
thanks for the tests we really need, saw similar results in my own testing.
What about if you were to disable e-cores in order to lower heat and give more room for overclocking? I wonder what that would look like. Honestly I dont think it would be worth it since the 14th gen CPU is likely not the bottleneck when gaming. thanks for the data friend.
I use this to run almost my entire OS on my last 8 e-cores and run my games/heavy apps on my 8 p-cores and 8 remaining e-cores. It’s been amazing and performance is much more stable, I’d recommend not fully making games use only p-cores nowadays as most can actually leverage them for background threads and other light threads the game needs. All in all, when used correctly Process Lasao is an awesome & powerful tool
So basically do not use process lasso? I am not that advanced to computing so please correct me
Would be interesting to see the same benchmark with disabled E cores through the BIOS. If there would be a difference between this method and Lasso method.
Also as other s have mentioned, would be awesome to see the same testing on non E core Intel CPUs and on AMD platform.
Otherwise, really great benchmarks!
Thanks for the video!
I mean he did a massive E cores off vs on in bios benchmark run with 40 games, as he noted on average disabling e cores even through the bios drew a slight performance regression, with few wins here and there on individual benchmarks.
I have a full amd build and play tarkov so here is my two cents. Process lasso can help with single core performance biased titles (such as tarkov) *IF* your AMD cpu is a little older. the newer ones (i upgraded recently) have a lot better single core performance in general stock out de box so its not really needed. Oh and the bitsum power plan, probalance etc caused me weird situational stutters (flicking too fast, 180's and scoping in) but when i changed back to the regular high performance option and no probalance etc the stutters dissapeared. I assume this is subjective though as every setup is different and will have different qwerks but thats my two cents.
You can already disable E-Cores and set process affinity through Window's Task Manager (Details tab). If Process Lasso has issues with Intel CPUs I would assume Windows' Task Manager would do a better job.
I have a feeling this program is interfering with the Memory Reference Instruction by eliminating both AMD'S,and INTELS own set Instruction utilizing cores. I would definitely try the same program with non E/P core on Intel and likewise to deduce the issue here. Also I would think higher end R.A.M. Cards could make up for that drop with a good CL. I really loved this video Brother, as it once again proves that we should never go on "hearsay". Windows has alot of work to do
That was a great point about not blindly following or believing what ppl say on forums or CZcams comments. Take everything with a grain of salt and if you can Always! Always test it yourself.
Thats right don't beleive everything thats why i checked agains many forums then test myself based on multiple posts.
Your wrong process lasso works amazing 9h my.gamijng..almost double fps..so ima stick with it and I hate to say this but if majority say it helps.and one who doesn't who are u gonna trust..u must be a conspiracy theorist too..what is the earth flat. As. Well??
@@Shuttterbugg Imagine being this stupid in real life. God bless this individual. ^
@@Shuttterbugg Im not going to believe any of them. Im gonna do my own independent research. You definitely sound like someone who would've freely drunk the kool-aid or took the stabby. Taking everything with a grain of salt and doing your own independent research is the only way to ensure you are making the best decision for you. If you are someone who follows the masses. You are simply a Sheep "I hate to say this but if the majority say it helps." what kind of dumb statement is this? I dont care what the majority says. Just because you are part of the majority doesn't mean you are correct. GTFOH. You are definitely a Sheep.
If process lasso doubled your performance, maybe consider throwing the piece of shit your use as a PC out and buy something that is at least not 30 years old lol.@@Shuttterbugg
Idk about amd but with intel in Windows Task Manager you can assign core affinities and specific cores to tasks just like process lasso
Premise: I am not a process lasso user, I am watching this video because I wanted to know more about the program and maybe consider using it. Unfortunately this video shows nothing.
1. All you did with process lasso is changing cpu affinity (in this case the program selected p-cores and deselected e-cores automatically) something you can do from the stock task manager manually selecting the process and then selecting the cores you want.
2. Process lasso has way more features than cpu affinity (that, again, it's something you can do from stock task manager, it's not a feature it's just a matter of comfort to have everything in one software).
3. Some people are pointing out that the recent intel architecture doesn't work well with setting cpu affinity like you did.
I'll be honest this is the first time I've heard people recommending Process Lasso for a performance boost, as if they are living under a rock. I use Process Lasso for a different reason for better hardware sleep control, idle power consumption control and seeing activities of left overs from potential bloat applications like Anti Cheats, anti tampering tools, etc.
But I didn't get to own a RYZEN 3000 CPU so that may have been "right" for them considering that ZEN2 had bad schedulers for early Windows 10 builds at some point. From Ryzen 5 2600 for 6 years jumping to Ryzen 5 5900X then downgrading to 5600 non X.
I've also heard similar sentiment from people using modern CPUs whether its Intel or AMD that you do not need Process Lasso to boost FPS. Still the tool has its use if you want its other functions especially in security reasons.
Using project lasso the correct way gives a good boost in synthetic scores esp cinebench r23. Esp on amd chips
Is it fine to just use this program to permanently set priorities and disabling cores than using any of that "performance mode" and "probalance" bs
I manually asign all windows applications to the E-cores and manually assign all P-Cores to my games, haven't seen numbers like yours to be honest, most games run about the same without plasso, but never a huge regression like you show, some games like Starfield run the same but with slower loads because of less available threads so i leave that game alone. But in all honesty the best idea is to assign all P-cores without the HT part to the games that find a benefit from it (most Single Threaded games) and leave multy threaded games on default.
"Just Lasso Your OS Bro." Hahahaha I thought about Jufes from Frame Chaser when I saw that.
Same here 😂
You tested a Ryzen 7950x riiiight? Process lasso works better on Ryzen NOT intel I have been doing tests on it since Zen2. And did some tests on Zen3 and use it on Zen4 RN. Zen3 and Zen4 is most noticable. It works because Ryzen Infinity fabric has a latency penalty between CCDs. Removing that fabric from the game improves performance. I also use my 2nd CCD to stream on Twitch so separating the 2 CCDs is beneficial so one Program Doesnt get Slowdown. Kinda like having 2 CPUs in one lol. Intel is Different Design. The way the cores communicate is different. But it does Work best on AMD. So if you don't stream and were to assign alot of the windows background processes to CCD2 and the game to CCD1 you actually get more FPS on AMD. The games that don't get as much improvement are the ones that need more cache cause removing the 2nd CCD halves your CPU cache tecnically.
Metro Exodus clearly had an incompatibility with the CPU affinity rule you set. Including that 0 FPS reading in the 12 game average accounts for almost all the difference. Most games showed no significant change. Process Lasso did exactly what it was designed to do: Give you control of running processes.
So Intel isn't just saying there is more going on under the hood with APO than simply forcing apps to run on P cores, I guess it makes sense given the difference between intel's hybrid architecture + thread director vs. AMD's chiplet implementation. Makes me wonder if games are optimized at all to use thread director in the first place and how much of a difference Windows 10 vs. 11 makes.
Ofc they are that's why it's so limited. Apo is good at leveraging the ecores for games actually when they're needed that's why the uplift is so large. If you weren't aware when ecores are talking to different clusters their latency is only like 20% worse than pcore to pcore. That and then they can get all the l2 cache capacity and bandwidth to themselves making them significantly higher ipc when only one of each cluster is being used.
Only improvement I've seen is when running cinebench r23 when setting the CB process to realtime
I have a 7950x3D now. Probably the processor that Lasso could benefit the most from my understanding. Truth is though, it just isn't needed. When the processor first launched and drivers where not the greatest, and it was fairly beneficial. Now all process loads go to the normal cores unless it needs more than 8 cores and 8 threads. It will ramp up the other V-cache cores if needed or if I boot up a game or manually tell it to use the V-cache cores first with Gamebar. Once you assign a process to the V-cache it will stay that way until you tell it otherwise through gamebar.
7950x3D works fantastic without Lasso now that the drivers have been fixed, and I haven't found a use case for it anymore.
Edit: I will tell you this Last of Us will eat through more than the normal 8 cores 8 threads, which may be why disabling the E cores has such an effect. It is one of the few games where more cores can mean better frames iirc.
I have a 7950X3D myself, and I only partially agree. I got rid of Process Lasso simply because I had small issues here and there with it and it was too much work. But default I could see that during gaming the wrong chiplet was sometimes accessed for a very brief moment, introducing bad 0.1% and 1% lows. For now I have set the non Vcache chiplet to 5050 max so that the system will always use the VCache chiplet first, and it removes the brief spikes downward in games.
@@nossy232323 Might depend on the game but I haven't noticed issues with core parking. I am not being overly watchful though either anymore.
It's not going to help when you have a lot of CPU headroom available. If you're at the limits of your hardware, it definitely helps to prioritize the game you're running to reduce stuttering and frame drops.
This has helped me keep a 4690k/gtx970 system going, with quite a bit of manual configuration and lowering priority of unimportant processes. Without it, I can't watch a stream/YT video and play certain games without one or the other stuttering. Some of this can be done in task manager, but that is not persistent through a reboot, while Process Lasso will automatically apply my configuration and allows for much more advanced tweaking.
I spent a lot of time figuring out what works best for this system over the years, YMMV. I can't imagine it being nearly as useful for higher end, current-gen CPUs. Priority settings have been much more important for me than affinity. It's not always a "push button, get performance" type of software, as much as they'd like you to believe that. It can be incredibly useful, depending on the system and configuration.
This will be helpful for the 5% of gamers who can afford a similar high end pc setup.
"its unclear why are some games behaving like there" because many of modern games at launch check what cpu you have and how many cores and save it in temporary settings(usually documents like Call of duty) and while in middle of game that launched with E cores you forcefully not allow it to use by 3rd party program while game still assume you have them enabled
Props to intel, looks like thread director works better than a diy core selector.
D, the best way to increase performance in a sim among other core aware apps is to use the e cores this has nothing to do with multitasking and a SIMD (single instruction multiple data) processing element and or that is a parallel array. So if a game has been optimized this way, compiled this way and the e cores get turned off guess what. Good report your method discoveries the thread directed v non thread directed optimized titles for their revision. mb
Or at least the games compiled for e cores tasks that can be substantial think a mini GPU in 4/6/8 core processing array. mb
its 2D x:y (not z) but GPUS do this. mb
Mbwhoiscool?,that you bro?
The only time I saw Process Lasso done any good is on older games and systems. And I am talking like Battlefield 2 type of games. Basically the games that isn't quite optimized for the OS and some older gen system that help with scheduling of the CPU. On older system (older HP laptop with like old AMD CPU GPU combo) I remembered Stalker Anomaly (basically one - two core DX 8 to DX 11 game) taking the same most powerful CPU which is also taken up by other background tasks. I think Windows Game mode and other optimizations came along the way with Windows 10 and the improvement of Hardware just neglect the need for Process Lasso.
The only good use process lasso has to me is cpu priority. Its also recommended by the devs to lower cpu priority for other apps instead of increasing it for your games. Also i tested the cpu affinity and i would not recommend messing with that its inconsistent.
CPU affinity is not very consistent approach, CPU sets in process Lasso is better as it does not prohibit the CPU to what cores to use explicitly more just gives the process a hint on what to use, but will still allow the CPU to direct it. Which I guess is where the lag is coming from when setting CPU affinity manually alongside any other solution by Intel or AMD under the hood.
Just to add I think what is desired is better scheduling from Microsoft to handle chiplet designed CPU's.
I have never seen this app recommended for improving gaming performance for non AMD X3D CPU's. 3D Vcache cores are not the same as E-cores.
it actually help me personally
I tried Lasso on Witcher 3 Next Gen to see if it would help with it's poor CPU threading which kills Ray Tracing performance. I didn't really see any gains in performance that couldn't also be attributed to run to run variance and margin of error. Any gain was virtually imperceptible in game play and certainly not worth having to put up with that stupid nag screen and nor worth paying good money to get rid of the nag. I tried it on a few other games and got essentially the same results. My system is 5800X, 32 GB 3600 dual rank DDR and a 4070 Ti @ 1440p.
Process Lasso is really designed for optimizing Servers not gaming and it does work very well for optimizing servers in the right hands.
this is not a proces lasso test dude you only prove forcing p cores doest work good
Always enable one E core per cluster so that they can utilise the cache entirely. Atleast that's what the Intel PBO thing seems to be doing.
yup and core to core latency is massively better when the ecores are talking to different clusters. You can also see thread director load up different clusters first too, so they know this ofc and I think APO was actually leveraging them to a greater extent like this being essentially an additobal 4 rocket lakeish ipc cores that can pick up some work when needed. Yes the ecores do have much higher ipc alone in case u were wondering because that l2 bandwidth is also shared among the 4 cores too.
How do you do that with Process Lasso
@@_atinsy Just count lmfao. They're in order. If you have a 13900k pick one from ecores 1-4, one from ecores 5-8 etc.
If I get it right, Intel has made 14th gen exclusivity performance boost by some HW locked optimizations, while 14th gen cpus share same cores with 13th/12th.
Then some 3rd party made attempt to enable that performance boost on 12th and 13th gen cpus, showing boost on just 2 selected titles, because they found out those games favour certain settings and use probably some less used coding, thread ordering etc, idk).
But that P. L. app affects negatively all other commonly coded games.
Let's call the app with fitting name; Project -l-asso😅.
Keep doing well. Merry Christmas.
This is more likely somewhat subjective to Intel architecture. I'm using on AMD 7950x3D and it is amazing overall. It's not the best but does it's job for what it is intended and works in help of xbox solution provided by AMD. I for one tested with and without it and numbers are better with Process Lasso on for my use case and with the settings that I've done for my specific need (development, gaming and content consumption).
I will love you if you can give me the lasso processors to use for my 3900x. Retail wow the only game I can't run at stableish frames.
Tbh, I've been using Process Lasso on most of my rigs, (including the ROG Ally), and I don't really FW the settings and just leave them at default. It fixes some of the weird issues that I have on some of my games, (especially games that stutter or have random fps drops for no reason, like Persona 5 Royal), and doesn't really affect the other games in my library.
EDIT: I exclusively use AMD CPUs, so that's probably why it works super well for me.
Same for me.
Intel APO is coming to 13th Gen and 12th gen btw
I heard, great news! Glad to hear they listened to the feedback.
Process lasso help me get lower temps on my laptop which helped me get better fps
Man some of those 1% lows are worse than amd x3D chips when you move fast in game... 😅 process lasso gives intel amd dip 24/7.
Not the AMDip. 🤣🤣🤣🤣🤣
Dude,there is SO MUCH with process lasso then what u tested😅
Pointless doing all this when all you have to do is unpark the cores. All stutter is gone now.
Lol wait, they limited IPO to 14thgen... even though 14th gen is the exact same as 13th gen, just refreshed? Lol what
From my experience, process lasso can be useful for games that struggle a lot with e-cores enabled.
Using it to block sekiro from accessing e-cores eliminated stutters that I've experienced with my 13600k.
I also like that it sets the high performance mode when I start the games, and keeps my pc in power saving mode while browsing the webs.
Saving the planet 1W at a time.
But my own testing also showed that disabling e-cores in the bios is much better for performance.
I'm using pl, cuz I'm way too lazy to go into bios whenever I want to toggle e-cores on and off.
Because I also do a few workstation type tasks, that do benefit from e-cores.
HAMPTER IS KING
🐹👑
I Only use process Lasso on games like Arma 3
Anyone that plays mmos and has 5800x3d tried it ?
Bump
GRRRRRRRRRRRR JUST PUT ALL THE SERVICES AND CORE WINDOWS OS ON 4 CORES GRRRRRRRRRRR RAAAAAAAAHHHHHHHH, YOU GOOBER, YOU GET BETTER FPS! WAAAAAHHHH, I AM ANGRY, SO ANGRY RAH, WHAT THE HECK RAAAAAAAAAH!!!!!!! (jk) MAKE SURE THEY ARE ON LOW MEMORY & LOW I/O RRRRRRAAAAAAAAAHHHH
This wouldn't help on a shitty laptop would it
This is so stupid, i need to go to BIOS to play most of games smoothly! Intel's strategy is against customers.
13900k this program would make my game freeze for 2 seconds aka Tarkov program is trash