Jump to content

Ouiche

Member
  • Content count

    13
  • Joined

  • Last visited

About Ouiche

  • Rank
    Fireteam Leader
  1. Hi. I found something interesting messing with the threads affinity with my 8 core Ryzen. Deactivating most of the core lead to a small increase of FPS. 1/ Go to the shooting range, wait 30 sec for the textures to load and fps to stabilize, note the fps. 2/ Alt-tab, open the task manager, details, right click on Squad, set affinity. If you have an 8 core, deselect everything but the core 9, 11, 13, 15. Come back to the game 3/ Come back to the game, look at your fps. If your cpu react like mine, you should gain a couple of FPS. Now, it's complicated to assess how much fps you're going to gain in multiplayer game. If you have a Ryzen, could you test and tell me if you see something? (don't forget to avoid gpu bottleneck) -Why one out of two? To mitigate the negative effects of SMT. SMT can cause a small loss of performance in some games. Can't hurt. -But why not using all the core and have something like 0, 2, 4, 6, 8, 10, 12, 14? Ryzen it not made of 8 core with 16mb of L3, but 2 groups of 4 core (called CCX) with 8mb of L3 each. Each CCX is connected to a "data fabric", a sort of bus managing a lot of stuff at the same time. That bus have a small bandwidth, we're talking of 38 to 60go/s depending of your DDR speed. It's small, but it do not seems to be a bottleneck is most case... Most. In some apps, it's possible to have a loss of performance when said app is using both ccx (not at 100%, even a very small load is enough). It's not clear yet, but the performance loss is related to the way CCX access L3 and / or threads moving from one ccx to the other, causing a noticeable increase of L3 access latency. I can't explain how or why, my knowledge is quite limited on that matter and even specialized hardware testers struggle to understand exactly how it work. Somehow in squad, there is a small gain of performance when you limit the game to one CCX. I tested with SMT off and on, and i see a small gain of FPS in both case by using one ccx (squad don't even use . Not big, but it's free. Note : the cpu is overclocked and the frequency is stable no matter the number of core used, the gain can't be caused by some kind of frequency boost. If you got a Ryzen, can you test?
  2. Too pixelated

    Try supersampling x1.25, txaa, sharpening at 1. I get great results with it and you should not really lose fps in multiplayer game. (i'm assuming you got the same fps in multiplayer game no matter the graphic setting?). You may lose fps in the shooting range but it's normal, it's less heavy on the cpu that multiplayer game and do not reflect "real" performance.
  3. Help me choose a monitor.

    Looks like a good computer. 6700k is strong for gaming, 16gb of ddr is just what you need, SSD for system + a couple of games and HDD for data is usually the way to go... and the 1070 is a very good card (i got one, love it). Nothing much to say. For the monitor, 144hz is a clear plus (even if you can't max it in my opinion), and gsync sound great... but it's also more expensive. Gaming IPS panels looks great, but they're very expensive and there is a lot of horror story about them (dead pixel / dust / light bleed), for the moment I'd rather go with a good / great TN panel. Dell got two interesting one - 2417DG (165hz, QHD in a 24' panel - ok, a bit weird - gsync, ~370€$) or the 2717dg (QHD in a 27' panel, 144hz, gsync, ~540$), both with good color out of the box for a gaming monitor and low response time... i'm thinking of buying one, but i just changed my CPU+MB+Ram and my wallet isn't very happy about it
  4. Anti-tank mine. It should be added to the loading screen picture, so far i placed 5 mines... and teamkilled two friendly vehicle :-|
  5. GTX1070 owners topic

    My bad, i failed to notice the difference. True, if he got vsync on if may explain the fps. Could you click on the interrogation mark in gpuZ (right side, next to PCIe 16x 3. @x16 1.1) then clic on "start render test"? The value "@x16 1.1" should change for something else, could you tell me what it is? The rest of the value reported looks fine for me. Gpu at 70c° is normal in gaming, i usually get around 40-80% of gpu utilization depending of the map / player count / action... and supersampling at x2 get me 50fps in the shooting range (and performances are far better here than in a normal game).
  6. GTX1070 owners topic

    30-60fps is too low for a 7700k, it's the most powerful cpu at the moment for single thread operation (and it's probably the most powerful overall for gaming). If i get ~60fps on average in 70-80 players servers and supersampling at 125% with a ryzen and a 1070 , you should at least get the same. Squad is limited by the cpu... and your cpu is better than mine for games like squad, arma2 / 3. The 1070 is more than enough to provide a solid 60fps in 1080p gaming without supersampling, as long as the cpu is not limiting the fps. What kind of fps to you get in the shooting range? Check the cpu / gpu max temperature (Use Hwinfo64 for example), if they're overheating it would be possible to have something throttling. Try to clean your client information (can help to solve fps issues), if it's not working roll back to previous GPU drivers (i was having weird resolution / scaling issue with the last nvidia driver, rolled back to 381.65 to fix it). Just in case, look at GPUZ and check the PCIE bus interface. It's possible to have your GPU pcie interface run in PCIE3 2x or even 1x, and it's going to destroy the performances. Is your GPU installed on the top or bottom pcie 3.0 slot? Do you have an M2 drive or anything connected to the sata express port? How the motherboard pcie slots are populated?
  7. Cpu bench in squad

    No oc (factory frequency), temperature is far from the limit (around~65-68c° in gaming, the card can go up to 82 without throttle) and the power limit... i don't think so, in short the Pascal GPU is designed to throttle constantly. (it automatically adjust / reduce the maximum frequency according to the maximum TDP, temperature, voltage and load, it's always hitting a limit). It's possible to increase that power limit but there is little to no benefit unless you're overclocking. I would be interested to see how other computers are doing on that matter.
  8. New Player Setup - Will it run?

    I think there is something wrong (unless you're trying to run the game in 4k). I get 50-70fps on 80 players map with a ryzen at 3.7ghz and a 1070 in 1080p... with supersampling at 125%, and the 7700k is performing better in gaming (especially when it come to games who require strong monothread power like squad). You should get better performances, try to clean the game cache and see if it improve the situation.
  9. New Player Setup - Will it run?

    It will run, but yes, overclock that I5. What kind of motherboard do you have? The 7400 may be difficult to overclock due to the lack of unlocked multiplicator. The 1070 is more than enough for 1080p / ultra (even with 125 or 150% supersampling) and can probably run the game in 1440p / ultra without issue, but the cpu may be short to reach 60fps on average without overclocking. if you haven't purchased yet, the R5 1600 may be a good alternative. It's a tiny bit more expensive but the overall performance is good / great and it's very easy to overclock. (it's not running very hot and it's sold with an ok / good cooler, you can push it to ~3.7ghz)
  10. Cpu bench in squad

    I used texture streaming yes. I deactivated SMT (amd hyperthreading) to see if it change something (and it's not really noticeable). For the thread swap, i have no idea how to record that accurately. I don't saw core changing the few time i checked. edit : all tests are done in 1080p, ultra, 125%. With texture streaming: Al Basrah - 74 players. Us side, start from FOB, calm until 1/4th of the game, heavy fighting in the town (north area) until the end (rocket spam, waves of soldiers, building to building fight) Without texture streaming (Al Basrah - 74 players. Insurgents side, joined mid-game, spawned in the town, random encounter moving into the buildings, long range fight trying to get close to apartments, had to fall back on refinery, heavy fighting - 20+ soldiers.) Overall, there was more action in the second recording and no driving time altering the average fps, but i think there is more spikes without texture streaming. Standing in the fire range, not moving at all, texture streaming on... And off. I still got spikes! There not as high / frequent as regular gameplay, but there is definitely something. Same firing range position, 200% No lag spike anymore. The fps are obviously lower, but they're also way more stable. I'm a bit confused. I was thinking the spikes could be caused by the gpu switching frequency / vcore due to the low use, but it do not seems to change at all under gpuZ (maybe it's so fast it's not recorded?)
  11. Hacking prevalence

    I can't recall 50% of the kills i do. I'm not a great player (ok-ish, i guess) but i played ArmA 2, 3, Battlefield 3/4/1 and... in squad, a good half of the kills are done by suppressing an area and getting lucky. SL yell "suppress bearing 121 NOW" so i empty a mag. I see smoke in the moutain, i shoot at the smoke. I see pixels moving in bushes and take fire from that area, 5-30 bullets to make the tree stop shooting at us. I don't pay the bullets. Writing that i realize my SL must see me as a kind of badly trained Labrador able to operate an ak74 I finished a game with 4 kills without clearly seeing a single enemy once. Someone getting 30+ kills on foot... that's fishy, but a good OG7 or M203 user can get a crazy amount of kill in a legit way (especially in town fight). I saw guys sitting on ammo box killing entire squads with 1-2 RPG, i think 20 or 25 kills is possible.
  12. Cpu bench in squad

    Hi, As a new player, i noticed a lots of questions regarding CPU performance and squad, but few reliable information available. Giving a rough estimate from an FPS counter in the corner of the screen without may not be the most accurate way to assess a CPU performance - it vary a lot between maps, player count, server, and it's actually not displaying everything. How to solve that? Fraps frametime benchmark + frapsbenchviewer. Frametime record each frame time in game. Frapsbenchviewer convert them into a graph. I don't think the bench is affecting the performances, i did not noticed any unusual performance loss. 1/ Use the benchmark option in fraps with the "frametime" option (work with the trial version) 2/ Once it's done, drag & drop the file in frapsbenchviewer 3/ Clic on "copy chart as image to clipboard if you want to save the picture. Note : -change the recording key for something else than "F11", the game freak out when i push that key for some reason. (goes from fullscreen to windowed, sometime the resolution change) -don't forget to start after spawning at the beginning of the game, and to stop as soon as possible after the game end, for a reliable result. (otherwise, you'll get 200+fps in the menu) -the graph show the frame time in ms. Lower = better. Spikes going up = bad. It will give you the average, the 1% low and 0.1% low. Average is not everything, the 1% and 0.1% low are very important for the overall fluidity (stuttering / micro lag spike are never good for that). Ex: Chora aas V2 with 72 players (few minutes of running since i join mid-game, lot of combat on the way and around insurgent checkpoint with an SVD, then we won by tickets near mosque) -Ryzen [email protected] -Gtx1070 -1080p, everything on max but the blur, 125%. My GPU is never fully utilized due to the CPU bottleneck (usually, ~70-80% max), that graph should display the raw cpu performance without any gpu limit. You can spot the time when i died and had to wait in the menu ("flat" line around 60k). I don't know what's causing the ~40ms frame, it seems to be a recurring event but i don't feel a lag. Could it be the map? I will try with full texture loading next time and see if it change something. It would be interesting to have various CPU tested, for those looking for an update. I'll record / post more benchs if you're interested. (but you can do it too, it's really easy). Of course, two game can't be compared but it can give people an general idea on how a cpu is going to perform in a real game. (clic on the picture for full resolution)
  13. Recommended cpu

    The 1600 could be interesting, but it's a little out of the budget (1600 : ~240$, MB : 90$, 16gb of ddr : ~140$) The 1600 come with the Ryzen 1700 cooler (so you save a couple of $), enough to overclock the cpu around ~3.7ghz (bit noisy in burn test, but mostly silent in game). You can get good MB for ~80-100$, but the ram is very expensive at the moment and i don't think it's going down anytime soon (~140$ for a 16gb kit). Sure, the 1600x get more base frequency, but it cost 40$ more and you have to buy a cooler (at least ~30$). As for squad, my Ryzen 1700 at 3.7ghz is getting ~50-80 depending of the map and number of player, most of the time 60 but it drop a bit on some maps. It's not the most powerful monocore wise, but the game feel fluid. I played ArmA 2 at ~25fps for hundred of hours back in the days, so a solid 50+ feel great (note: a 6 core ryzen is going to have exactly the same game performance as an 8 core one at the same frequency, the 2 extra core do nothing, even in battlefield 1 they're not used). That being said, for pure gaming performance it's hard to beat the 7700k (especially on "monocore" games) at the moment... but it's out of your budget. The 7600k is "meh", now that we can get 6 core ryzen. Lots of today's games benefit from more than 4 core (being hyper threading / SMT, or real 6 core) especially on minimum fps.
×