Page 49 of 165 FirstFirst ... 39 47 48 49 50 51 59 99 149 ... LastLast
Results 481 to 490 of 1649
  1. #481
    ROG Enthusiast Array
    Join Date
    Jul 2018
    Reputation
    11
    Posts
    44

    Quote Originally Posted by AlleyViper View Post
    That is very weird. PE A/1/2 that forces scalar 10x, more boost time and tdp margin, should help you mostly with >2 core loads with more time under higher boost, but not so much in single thread. My single core score is also about ~175 with a 2600x and 3200 c14 1T with PE2 (-0,1V vcore offset) and about ~1420 all core. With PE2 doesn't seem to go over 4250MHz in single core testing, but it could be probably because it's already over 60ºC on my TRUE with push+pull.

    I don't remember taking such hit in single thread without PE A/1/2/PBO, because XFR2 should still let you boost to 4250MHz without PBO. At that speed it should still be in the ~170 range, not 155.

    Did you test it with the balanced performance profile, to ensure other threads were kept at a low tdp? You might check if setting a negative offset will let you achieve a better score without PE for the same reason. Next week I'll be able to give you some numbers to compare without PE (I'm away from my AM4 ATM).
    No, everything on the motherboard is left as it is when you get it out of the box except DOCP to 3200MHz and BIOS update v4024. All I did was change the "Performance Enhancer". I've no intention on diving into changing specific XFR/PBO values but thanks for the offer. I prefer the "out of the box" experience as I believe it gives me the best performance already. Besides I already plan to sell this CPU once Ryzen 2nd gen comes out as I will be upgrading by then. Hoping 2nd gen is more overclockable just like my old 2700K @ 5GHz 1.5v.
    Last edited by dfg555; 11-17-2018 at 04:07 AM.

  2. #482
    ROG Guru: Blue Belt Array gupsterg PC Specs
    gupsterg PC Specs
    MotherboardAsus Maximus VII Ranger
    Processori5 4690K
    Memory (part number)HyperX Savage 2400MHz 16GB
    Graphics Card #1Sapphire R9 Fury X (1145/545 Custom ROM)
    MonitorAsus MG279Q
    Storage #1Samsung 840 Evo 250GB
    Storage #2HGST 2TB
    CPU CoolerThermalRight Archon SB-E X2
    CaseSilverstone Temjin 06
    Power SupplyCooler Master V850
    Headset/Speakers HyperX Cloud
    OS Win 7 Pro x64 / Win 10 Pro x64
    gupsterg's Avatar
    Join Date
    Feb 2015
    Reputation
    102
    Posts
    982

    Quote Originally Posted by gupsterg View Post
    Also read few pages back a member stating when changing BCLK that PB/XFR2 disables, if you guys have Core Performance Boost option in UEFI then set that to Enabled. Then PB/XF2 should stay active even with BCLK change.

    Quote Originally Posted by Rohnlex View Post
    How ? it only lets me select Auto/Disabled
    Ahh, UEFI gimping . I saw this with X370 as well, that Crosshair had the most UEFI options and as you went down the tiers things disappeared.

    On OCN there is a thread where UEFI modding allows options to be gained, perhaps there is UEFI for X470-F Gaming, dunno.

    Quote Originally Posted by Alex0915 View Post
    I think it's a fault of Core Temp I got the warning from it even when I set the threshold temp to 80C.

    I think a hot room was at fault my room temp was at 30C because of the heat so I opened a window and now the temps are in the 40-60 range. I have a 4 copper heat pipe cpu cooler and 4x 120mm 83 cfm case fans and a 2 copper heat pipe gpu cooler. The GPU temps are cooler than my CPU temps even with 3D progs running and the twin fans on it dont even usually spin.

    I hear my fans all spin up when temps exceed 60C.

    I wonder if the throttle temp thing works with this board?
    Please state model of CPU cooler, it really does seem as if this is the reason behind your excessive CPU temps.

    I use a lapped base ThermalRight Archon IB-E X2 with 2x TY-143 fans, room ambient for me is ~20-24C. I also have 2x TY143 as case intake fans, they use CPU header PWM so as CPU fans increase speed they do as well. I have 1x 120mm and 2x 90mm case exhaust fans.

    With PE: [Default] I see just over ~70C if I was loading CPU with P95.

    With PE: [Default], PBO: [Enabled] and BCLK tweak of 102.2, so single core is ~4.45GHz and multi averages ~4.15GHz in P95, I'll see max ~75C when loading with P95, depending on room ambient temp.

    Daily usage there is no chance I'd see the temps I see in P95 from what I have seen. I have used 2x 2700X on C6H/C7H. Below testing can get loud at moments, as stated before when CPU temp goes past 70C fans will go to 100% and the TY-143 can reach 2.5K RPM. The way my case is and where motherboard temp sensor is, it's usually inline or +2-3C higher than room temp. Below is ~8hrs P95 run and 1hr RealBench stress mode.

    Click image for larger version. 

Name:	P95 v28.10b1 28GB PBOE 102.2 +12.5mV 3200GT0.3 1.006 1.35 0.679 60 room 24C PASS 8hrs.jpg 
Views:	8 
Size:	865.9 KB 
ID:	77046

    Click image for larger version. 

Name:	RB PBOE 102.2 +12.5mV 3200GT0.3 1.006 1.35 0.679 60 room 22C WP PASS 60min.jpg 
Views:	6 
Size:	759.0 KB 
ID:	77048
    Last edited by gupsterg; 11-17-2018 at 09:29 AM.
    Intel Defector AMD Rebel

    TR 1950X - Custom WC - Asus Zenith Extreme Alpha - G.Skill Trident Z 3400MHz C14 QC - RX VEGA 64
    R5 3600 - Custom WC - Asus Crosshair VII Hero Wifi - G.Skill Ripjaws V 3800MHz C16 - Sapphire HD 5850 Toxic

    24/7 OC: i5 4690K @ 4.9GHz CPU@1.255v 4.4GHz Cache@1.10v - Archon SB-E X2 - Asus Maximus VII Ranger
    Sapphire Fury X (1145/545 ~17.7K GS 3DM FS)

  3. #483
    ROG Enthusiast Array
    Join Date
    Nov 2018
    Reputation
    10
    Posts
    55

    @dfg555 The only problem with leaving everything "Auto" is mostly auto voltages. Using DCOP for 3200 will push your VSOC to 1.15V, while something like 0,986-1.05V could be enough (just notice the caps above). I see no reason for forcing anything over 1.1V. PE Auto also sets PE1(or 2 at most) which forces scalar 10x, and makes your CPU draw too much voltage at higher speeds for a longer time, which makes a greater difference when more cores are pushed at a higher speed than stock PBO/XFR under load. Hence the need to lower cpu voltage offset by about -0.05 to -0,1V depending on what your CPU can take to ensure lower voltage spikes, slightly higher boosting and even better temps. This should also help stock without PE/PBO to gain some efficiency.

    @gupsterg for what Alex0915 says, I also really doubt about the capability of his CPU cooler. I don't remember any high performing tower with only 4 heatpipes, except the original Silver Arrow/IFX, so I'd guess an entry 12cm tower (Hyper 212, True Spirit...) or at best the older NH-U12P, all with short legs for a 2700X under avx2.
    I find my TRUE Black (the denser Rev.A 790g version), kryonaut, with 2 high pressure 1850rpm Gentle Typhoons at the limit for a 2600X (max 72º-73º on 8h of latest prime95, PE2 and -0,1V offset). It might perform roughly near a Archon with only one regular TY-141/147 fan at max speed, if not a tad better.
    Last edited by AlleyViper; 11-17-2018 at 10:01 PM.

  4. #484
    ROG Enthusiast Array
    Join Date
    Jul 2018
    Reputation
    11
    Posts
    44

    Question

    Quote Originally Posted by AlleyViper View Post
    @dfg555 The only problem with leaving everything "Auto" is mostly auto voltages. Using DCOP for 3200 will push your VSOC to 1.15V, while something like 0,986-1.05V could be enough (just notice the caps above). I see no reason for forcing anything over 1.1V. PE Auto also sets PE1(or 2 at most) which forces scalar 10x, and makes your CPU draw too much voltage at higher speeds for a longer time, which makes a greater difference when more cores are pushed at a higher speed than stock PBO/XFR under load. Hence the need to lower cpu voltage offset by about -0.05 to -0,1V depending on what your CPU can take to ensure lower voltage spikes, slightly higher boosting and even better temps. This should also help stock without PE/PBO to gain some efficiency.
    Well then I assume ASUS is to blame for this "Auto" setting then. Perhaps other motherboards don't do this then.

  5. #485
    ROG Enthusiast Array
    Join Date
    Nov 2018
    Reputation
    10
    Posts
    55

    On a friend's new Asus Z370 Prime running a 8700K, by default his board had on Auto a setting called something like "Multicore Enhancements". Guess what, it made all loaded cores boost to ~4.7GHz as if in single/dual-thread, instead of stock ~4.3ish GHz, causing about a 15-20ºC extra on heavy loads under his decent mid-range cooler that should have been enough for stock. It's a similar situation to this board's PE on Auto which applies a PBO overclock, but you'll probably only see up to ~50-150MHz gain on most multi-core loads with a 2600X, when compared to what can happen on a Intel CPU.

    If you check the latest roundups of 9900K reviews, most manufactures doing the same by boosting core speeds way above the original TDP limit (effectively overclocking) to get better performance out of the box. For now, at least Intel doesn't seem to bother about it to get better results on reviews. You can expect it to be the new norm, because no mb manufacturer would want to look subpar on out of the box reviews. Two links from techspot on this: link1 link2
    Last edited by AlleyViper; 11-17-2018 at 11:54 PM.

  6. #486
    ROG Guru: Yellow Belt Array
    Join Date
    Jul 2018
    Reputation
    10
    Posts
    193

    Unfortunately I have no idea what heatsink this is, I bought it as a motherboard-cpu combo to put in my computer case. All I can tell is that it's a Cooler Master brand fan and heatsink and the fan is smaller than my 4x 120mm case fans. And it spins at up to 6000 RPM when the temps get high. It's not a stock AMD set up, it has four copper heat pipes.

  7. #487
    ROG Guru: Yellow Belt Array
    Join Date
    Jul 2018
    Reputation
    10
    Posts
    193

    Quote Originally Posted by AlleyViper View Post
    On a friend's new Asus Z370 Prime running a 8700K, by default his board had on Auto a setting called something like "Multicore Enhancements". Guess what, it made all loaded cores boost to ~4.7GHz as if in single/dual-thread, instead of stock ~4.3ish GHz, causing about a 15-20ºC extra on heavy loads under his decent mid-range cooler that should have been enough for stock. It's a similar situation to this board's PE on Auto which applies a PBO overclock, but you'll probably only see up to ~50-150MHz gain on most multi-core loads with a 2600X, when compared to what can happen on a Intel CPU.

    If you check the latest roundups of 9900K reviews, most manufactures doing the same by boosting core speeds way above the original TDP limit (effectively overclocking) to get better performance out of the box. For now, at least Intel doesn't seem to bother about it to get better results on reviews. You can expect it to be the new norm, because no mb manufacturer would want to look subpar on out of the box reviews. Two links from techspot on this: link1 link2
    From what I've read on Tomshardware, liquid cooling is a must on those new Intel cpus. Both the 9900K and the 9700K.

  8. #488
    ROG Guru: Yellow Belt Array
    Join Date
    Jul 2018
    Reputation
    10
    Posts
    193

    Quote Originally Posted by AlleyViper View Post
    @dfg555 The only problem with leaving everything "Auto" is mostly auto voltages. Using DCOP for 3200 will push your VSOC to 1.15V, while something like 0,986-1.05V could be enough (just notice the caps above). I see no reason for forcing anything over 1.1V. PE Auto also sets PE1(or 2 at most) which forces scalar 10x, and makes your CPU draw too much voltage at higher speeds for a longer time, which makes a greater difference when more cores are pushed at a higher speed than stock PBO/XFR under load. Hence the need to lower cpu voltage offset by about -0.05 to -0,1V depending on what your CPU can take to ensure lower voltage spikes, slightly higher boosting and even better temps. This should also help stock without PE/PBO to gain some efficiency.

    @gupsterg for what Alex0915 says, I also really doubt about the capability of his CPU cooler. I don't remember any high performing tower with only 4 heatpipes, except the original Silver Arrow/IFX, so I'd guess an entry 12cm tower (Hyper 212, True Spirit...) or at best the older NH-U12P, all with short legs for a 2700X under avx2.
    I find my TRUE Black (the denser Rev.A 790g version), kryonaut, with 2 high pressure 1850rpm Gentle Typhoons at the limit for a 2600X (max 72º-73º on 8h of latest prime95, PE2 and -0,1V offset). It might perform roughly near a Archon with only one regular TY-141/147 fan at max speed, if not a tad better.
    It's Cooler Master so it could be the Hyper 212. I guess I'll have to keep the a/c on in the computer room because that is a surefire way to reduce CPU temps lol. The a/c faces the computer and I notice an immediate 20C drop in temps when I turn it on. I just have to keep room temps around 25C.

    At what temperatures does part longevity become shortened would you say? Anything above 70C? Today it's nice and cool in my room so no warnings lol. I'm thinking of lowering voltage like you wrote in your above post to lower temps but dont know if thats necessary if I can keep the temps under 70C. Do lower voltage spikes make the parts last longer?

    It sounds like you dont favor Auto settings, but is setting memory speed to Auto a good setting to maximize stability and keep temps manageable? There is no "default" memory setting, unless it's the specifically rated speed for the memory.
    Last edited by Alex0915; 11-18-2018 at 12:32 AM.

  9. #489
    ROG Guru: Yellow Belt Array
    Join Date
    Jul 2018
    Reputation
    10
    Posts
    193

    By the way, are Ryzen Timing Checker and Real Bench worth getting? Oh nevermind I already have Ryzen Timing Checker, I just haven't used it yet. What is Real Bench?
    Last edited by Alex0915; 11-18-2018 at 01:20 AM.

  10. #490
    ROG Enthusiast Array
    Join Date
    Jul 2018
    Reputation
    11
    Posts
    44

    Quote Originally Posted by AlleyViper View Post
    On a friend's new Asus Z370 Prime running a 8700K, by default his board had on Auto a setting called something like "Multicore Enhancements". Guess what, it made all loaded cores boost to ~4.7GHz as if in single/dual-thread, instead of stock ~4.3ish GHz, causing about a 15-20ºC extra on heavy loads under his decent mid-range cooler that should have been enough for stock. It's a similar situation to this board's PE on Auto which applies a PBO overclock, but you'll probably only see up to ~50-150MHz gain on most multi-core loads with a 2600X, when compared to what can happen on a Intel CPU.

    If you check the latest roundups of 9900K reviews, most manufactures doing the same by boosting core speeds way above the original TDP limit (effectively overclocking) to get better performance out of the box. For now, at least Intel doesn't seem to bother about it to get better results on reviews. You can expect it to be the new norm, because no mb manufacturer would want to look subpar on out of the box reviews. Two links from techspot on this: link1 link2

    Yea I already know about the whole TDP Intel/Motherboard agenda. Fishy stuff.

Page 49 of 165 FirstFirst ... 39 47 48 49 50 51 59 99 149 ... LastLast

Posting Permissions

  • You may not post new threads
  • You may not post replies
  • You may not post attachments
  • You may not edit your posts
  •