• Welcome to TechPowerUp Forums, Guest! Please check out our forum guidelines for info related to our community.

Variations beyween 7800X3D CPUs

Joined
Aug 9, 2019
Messages
1,567 (0.87/day)
Processor 7800X3D 2x16GB CO
Motherboard Asrock B650m HDV
Cooling Peerless Assassin SE
Memory 2x16GB DR A-die@6000c30 tuned
Video Card(s) Asus 4070 dual OC 2610@915mv
Storage WD blue 1TB nvme
Display(s) Lenovo G24-10 144Hz
Case Corsair D4000 Airflow
Power Supply EVGA GQ 650W
Software Windows 10 home 64
Benchmark Scores Superposition 8k 5267 Aida64 58.5ns
So after building 4 setups with 7800X3D I find the variations in imc/fclk/CO fascinating:
All cpus are tested with Asrock B650m hdv bios 2.08. I used different psus, all gold 550-650W and 3 different aircoolers. The following actual SOC/VDDP is required for M-die 6000c30-35-35 480 rfc errorfree in TM5 with tight timings and best performing fclk:
Sample 1: 1.11v/1.0v 2167
Sample 2: 1.28v/1.12v 2133
Sample 3: 1.02v/0.95v (1.07v required for A-die 16GB DR ram) 2200
Sample 4: 1.02v/0.98v 2200

As for CO, I say avg negative CO achieved. This is tested with occt avx2 extreme. Daily stable for most cases is probably 2-5 negative better :)
Sample 1: -26
Sample 2: -41
Sample 3: -35
Sample 4: -28

I currently use sample 3, the others sit in different computers I build for friends. The imc in sample 2 is si bad that it needs mote SOC running 5600 (1.04v)than sample 3 and 4 needs for 6000. CO of sample 2 is really golden. Several cores runs -50 stable.

All cpus can run vddio at 1.2v and below with SR, the one using DR needs 1.24v vddio. All samples had to lower fclk above 1.25v SOC.

Silicon lottery varirs greatly ;)
 
Last edited:
Joined
Jan 29, 2023
Messages
989 (1.86/day)
System Name KLM
Processor 7800X3D
Motherboard B-650E-E Strix
Cooling Arctic Cooling III 280
Memory 16x2 Fury Renegade 6000-32
Video Card(s) 4070-ti PNY
Storage 512+512+1+2+2+2+2+6+500+256+4+4+4
Display(s) VA 32" 4K@60 - OLED 27" 2K@240
Case 4000D Airflow
Audio Device(s) Edifier 1280Ts
Power Supply Shift 1000
Mouse 502 Hero
Keyboard K68
Software EMDB
Benchmark Scores 0>1000
Is there ID/specs i can find that tells me which i have like CPU-Z infos ?.. or should i go BIOS and set the "negative CO" ?.. or just look at DOCP enabled SOC/VDDP ?
 

tabascosauz

Moderator
Supporter
Staff member
Joined
Jun 24, 2015
Messages
7,875 (2.38/day)
Location
Western Canada
System Name ab┃ob
Processor 7800X3D┃5800X3D
Motherboard B650E PG-ITX┃X570 Impact
Cooling NH-U12A + T30┃AXP120-x67
Memory 64GB 6400CL32┃32GB 3600CL14
Video Card(s) RTX 4070 Ti Eagle┃RTX A2000
Storage 8TB of SSDs┃1TB SN550
Case Caselabs S3┃Lazer3D HT5
So after building 4 setups with 7800X3D I find the variations in imc/fclk/CO fascinating:
The following actual SOC/VDDP is required for M-die 6000c30-35-35 480 rfc errorfree in TM5 with tight timings and best performing fclk:
Sample 1: 1.11v/1.0v 2167
Sample 2: 1.28v/1.12v 2133
Sample 3: 1.02v/0.95v (1.07v required for A-die 16GB DR ram) 2200
Sample 4: 1.02v/0.98v 2200

As for CO, I say avg negative CO achieved
Sample 1: -26
Sample 2: -41
Sample 3: -35
Sample 4: -28

I currently use sample 3, the others sit in different computers I build for friends. The imc in sample 2 is si bad that it needs mote SOC running 5600 (1.04v)than sample 3 and 4 needs for 6000. CO of sample 2 is really golden. Several cores runs -50 stable.

All cpus can run vddio at 1.2v and below with SR, the one using DR needs 1.24v vddio. All samples had to lower fclk above 1.25v SOC.

Silicon lottery varirs greatly ;)

1.28V? Ouch!

CO variation seems to be significant for 7800X3D. Mine sits at -28 avg after Hydra but Hydra testing is pretty conservative and penalizes very heavily for crashes, -35 was fine for daily. So not that great, but I have seen some even unluckier people in 1usmus discord unfortunately. SOC not the worst at 1.0-1.05V.
 
Joined
Aug 9, 2019
Messages
1,567 (0.87/day)
Processor 7800X3D 2x16GB CO
Motherboard Asrock B650m HDV
Cooling Peerless Assassin SE
Memory 2x16GB DR A-die@6000c30 tuned
Video Card(s) Asus 4070 dual OC 2610@915mv
Storage WD blue 1TB nvme
Display(s) Lenovo G24-10 144Hz
Case Corsair D4000 Airflow
Power Supply EVGA GQ 650W
Software Windows 10 home 64
Benchmark Scores Superposition 8k 5267 Aida64 58.5ns
Is there ID/specs i can find that tells me which i have like CPU-Z infos ?.. or should i go BIOS and set the "negative CO" ?.. or just look at DOCP enabled SOC/VDDP ?
You just have to test. As simple and conplicated as that :)
 
Joined
Jan 29, 2023
Messages
989 (1.86/day)
System Name KLM
Processor 7800X3D
Motherboard B-650E-E Strix
Cooling Arctic Cooling III 280
Memory 16x2 Fury Renegade 6000-32
Video Card(s) 4070-ti PNY
Storage 512+512+1+2+2+2+2+6+500+256+4+4+4
Display(s) VA 32" 4K@60 - OLED 27" 2K@240
Case 4000D Airflow
Audio Device(s) Edifier 1280Ts
Power Supply Shift 1000
Mouse 502 Hero
Keyboard K68
Software EMDB
Benchmark Scores 0>1000
HWinfo stock & DOCP Asus Tweaked (better).

B650E-E actaul non-9000 BIOS (same date out btw).

Can you tell ?
 

Attachments

  • Presse-papier_01a.png
    Presse-papier_01a.png
    263.2 KB · Views: 36
  • Presse-papier_01b.png
    Presse-papier_01b.png
    28.9 KB · Views: 37
  • Presse-papier_09a.png
    Presse-papier_09a.png
    267.2 KB · Views: 28
  • Presse-papier_09b.png
    Presse-papier_09b.png
    29 KB · Views: 36
Joined
Aug 9, 2019
Messages
1,567 (0.87/day)
Processor 7800X3D 2x16GB CO
Motherboard Asrock B650m HDV
Cooling Peerless Assassin SE
Memory 2x16GB DR A-die@6000c30 tuned
Video Card(s) Asus 4070 dual OC 2610@915mv
Storage WD blue 1TB nvme
Display(s) Lenovo G24-10 144Hz
Case Corsair D4000 Airflow
Power Supply EVGA GQ 650W
Software Windows 10 home 64
Benchmark Scores Superposition 8k 5267 Aida64 58.5ns
HWinfo stock & DOCP Asus Tweaked (better).

B650E-E actaul non-9000 BIOS (same date out btw).

Can you tell ?
No. You have to test Testmem5 usmus config. Try lowering soc and vddp till you get errors. As for CO try occt avx2 extreme.
 
Joined
Nov 18, 2010
Messages
7,266 (1.46/day)
Location
Rīga, Latvia
System Name HELLSTAR
Processor AMD RYZEN 9 5950X
Motherboard ASUS Strix X570-E
Cooling 2x 360 + 280 rads. 3x Gentle Typhoons, 3x Phanteks T30, 2x TT T140 . EK-Quantum Momentum Monoblock.
Memory 4x8GB G.SKILL Trident Z RGB F4-4133C19D-16GTZR 14-16-12-30-44
Video Card(s) Sapphire Pulse RX 7900XTX. Water block. Crossflashed.
Storage Optane 900P[Fedora] + WD BLACK SN850X 4TB + 750 EVO 500GB + 1TB 980PRO+SN560 1TB(W11)
Display(s) Philips PHL BDM3270 + Acer XV242Y
Case Lian Li O11 Dynamic EVO
Audio Device(s) SMSL RAW-MDA1 DAC
Power Supply Fractal Design Newton R3 1000W
Mouse Razer Basilisk
Keyboard Razer BlackWidow V3 - Yellow Switch
Software FEDORA 40
We should consider all those number only when taking account vdroop and specific motherboard configurations where this number really differs. Your 1.2V are not the same 1.2V at my board in reality during specific stress tests.

There is no way to measure the summed voltage at the core as you cannot put probes there, but those numbers should not be taken as carved in stone, they are within ~10% tolerance, and that's still a lot. Basically it is chatter about nothing really. Each specific case is on their own rules and scenarios.
 
Joined
Jan 29, 2023
Messages
989 (1.86/day)
System Name KLM
Processor 7800X3D
Motherboard B-650E-E Strix
Cooling Arctic Cooling III 280
Memory 16x2 Fury Renegade 6000-32
Video Card(s) 4070-ti PNY
Storage 512+512+1+2+2+2+2+6+500+256+4+4+4
Display(s) VA 32" 4K@60 - OLED 27" 2K@240
Case 4000D Airflow
Audio Device(s) Edifier 1280Ts
Power Supply Shift 1000
Mouse 502 Hero
Keyboard K68
Software EMDB
Benchmark Scores 0>1000
I will have to know what i do first going BIOS:

SOC is the voltage that has been burning x3d CPU right, i know where it is...

VDDP, idk, will find, it's in same list, should be near SOC ?

CO ? that's CPU, it's in advanced settings right ?
 

Attachments

  • Presse-papier_A.png
    Presse-papier_A.png
    105.6 KB · Views: 21
Joined
May 2, 2022
Messages
1,355 (1.69/day)
Location
G-City, UK
System Name My first new build
Processor Ryzen 7 5700X
Motherboard MSI B450M mortar max
Cooling Deepcool AK620
Memory 32gb Corsair Vengeance LPX 3200@XMP3200
Video Card(s) Sapphire 6900XT Toxic Air Cooled
Storage Samsung 870 QVO 1TB x 2, Lexar 256 GB, Western Digital SN550 500 GB nvme
Display(s) 32" LG GN600-B
Case Aerocool Eclipse
Audio Device(s) Corsair HS70 wireless
Power Supply MSI MPG AGF 850 watt gold
Mouse Glorious Model D l Pad GameSir G7 SE
Keyboard Redragon Vara K551P
Software Windows 11
So I've purchased a 7800X3D and 6000 CL30 trident z. Can't I just put it all together, enable expo and be fine?

I'm asking as I have everything except the ram which is coming from the USofA.
 
Joined
Aug 9, 2019
Messages
1,567 (0.87/day)
Processor 7800X3D 2x16GB CO
Motherboard Asrock B650m HDV
Cooling Peerless Assassin SE
Memory 2x16GB DR A-die@6000c30 tuned
Video Card(s) Asus 4070 dual OC 2610@915mv
Storage WD blue 1TB nvme
Display(s) Lenovo G24-10 144Hz
Case Corsair D4000 Airflow
Power Supply EVGA GQ 650W
Software Windows 10 home 64
Benchmark Scores Superposition 8k 5267 Aida64 58.5ns
We should consider all those number only when taking account vdroop and specific motherboard configurations where this number really differs. Your 1.2V are not the same 1.2V at my board in reality during specific stress tests.

There is no way to measure the summed voltage at the core as you cannot put probes there, but those numbers should not be taken as carved in stone, they are within ~10% tolerance, and that's still a lot. Basically it is chatter about nothing really. Each specific case is on their own rules and scenarios.
Yeah, they will always be board specific, but a 260mv variance in required voltage for 6000 will be quite similar on other boards. CO will vary more as you say due to vdrop, but a 15 variance in CO will be noticable especially if cooler is avg.
 
Joined
May 22, 2024
Messages
187 (3.46/day)
System Name Kuro
Processor AMD Ryzen 7 7800X3D@65W
Motherboard MSI MAG B650 Tomahawk WiFi
Cooling Thermalright Phantom Spirit 120 EVO
Memory Corsair DDR5 6000C30 2x48GB (Hynix M)@6000 30-36-36-48 1.36V
Video Card(s) PNY XLR8 RTX 4070 Ti SUPER 16G@200W
Storage Crucial T500 2TB + WD Blue 8TB
Case Lian Li LANCOOL 216
Audio Device(s) Sound Blaster AE-7
Power Supply MSI MPG A850G
Software Ubuntu 24.04 LTS + Windows 10 Home Build 19045
Benchmark Scores 17761 C23 Multi@65W
I thought speeding fclk out of sync beyond 2:3 with uclk/mclk is actually counterproductive for latency and performance?

But note that's coming from someone who drew the short stick on silicon lottery, whose 7800X3D won't take any negative CO at all without errors: I wasn't able to stabilize it at all-core -10, and did not bother after that. I tested with y-cruncher and Prime95.

It did take 1.1V vsoc /1.05V vddp from the default, and set vddio at 1.25V when the dual-rank M-die is set to 1.36v, a bit less from the A-XMP defaults. 500 tRFC is as little as it could take.
 
Joined
Nov 18, 2010
Messages
7,266 (1.46/day)
Location
Rīga, Latvia
System Name HELLSTAR
Processor AMD RYZEN 9 5950X
Motherboard ASUS Strix X570-E
Cooling 2x 360 + 280 rads. 3x Gentle Typhoons, 3x Phanteks T30, 2x TT T140 . EK-Quantum Momentum Monoblock.
Memory 4x8GB G.SKILL Trident Z RGB F4-4133C19D-16GTZR 14-16-12-30-44
Video Card(s) Sapphire Pulse RX 7900XTX. Water block. Crossflashed.
Storage Optane 900P[Fedora] + WD BLACK SN850X 4TB + 750 EVO 500GB + 1TB 980PRO+SN560 1TB(W11)
Display(s) Philips PHL BDM3270 + Acer XV242Y
Case Lian Li O11 Dynamic EVO
Audio Device(s) SMSL RAW-MDA1 DAC
Power Supply Fractal Design Newton R3 1000W
Mouse Razer Basilisk
Keyboard Razer BlackWidow V3 - Yellow Switch
Software FEDORA 40
Yeah, they will always be board specific, but a 260mv variance in required voltage for 6000 will be quite similar on other boards. CO will vary more as you say due to vdrop, but a 15 variance in CO will be noticable especially if cooler is avg.

Sure. But you cannot apply X to Y here. If the board has a tendency to overshoot and supply more voltage actually, then CO numbers will be greater, you have more voltage budget. Just treat it as a knob and steps. It actually does not matter as long it performs the same.

You have to put the CPUs in the same system and force same temps to pick up a golden sample. You did it across multiple machines, thus the numbers are not comparable really.

Yeah, you can do some insight for the chatter, but it is all pretty random flavored to make some statements.

I have to agree that the tier levels is still a thing, anywhere silicon related. Even for phone CPUs and we have around 10levels. So in reality you can have same phone, but it can have 10 steps down worse voltage table, depending how lucky you are. But there you have no means to alter anything. So thank PC for being PC not Mac, and we still have the power to tinker around.
 
Joined
Jan 29, 2023
Messages
989 (1.86/day)
System Name KLM
Processor 7800X3D
Motherboard B-650E-E Strix
Cooling Arctic Cooling III 280
Memory 16x2 Fury Renegade 6000-32
Video Card(s) 4070-ti PNY
Storage 512+512+1+2+2+2+2+6+500+256+4+4+4
Display(s) VA 32" 4K@60 - OLED 27" 2K@240
Case 4000D Airflow
Audio Device(s) Edifier 1280Ts
Power Supply Shift 1000
Mouse 502 Hero
Keyboard K68
Software EMDB
Benchmark Scores 0>1000
No. You have to test Testmem5 usmus config. Try lowering soc and vddp till you get errors. As for CO try occt avx2 extreme.

Should i set SOC to 1.200 when DOCP sets it to 1.25* (instead 1.02* stock) ?.. and call it a day, bc i won't do the CO things anyway (i don't see VDDP in my HWi screens) !
 

Attachments

  • 20240707_142550.jpg
    20240707_142550.jpg
    583.8 KB · Views: 19
  • 20240707_142637.jpg
    20240707_142637.jpg
    2.2 MB · Views: 18
Joined
Aug 9, 2019
Messages
1,567 (0.87/day)
Processor 7800X3D 2x16GB CO
Motherboard Asrock B650m HDV
Cooling Peerless Assassin SE
Memory 2x16GB DR A-die@6000c30 tuned
Video Card(s) Asus 4070 dual OC 2610@915mv
Storage WD blue 1TB nvme
Display(s) Lenovo G24-10 144Hz
Case Corsair D4000 Airflow
Power Supply EVGA GQ 650W
Software Windows 10 home 64
Benchmark Scores Superposition 8k 5267 Aida64 58.5ns
Should i set SOC to 1.200 when DOCP sets it to 1.25* (instead 1.02* stock) ?.. and call it a day, bc i won't do the CO things anyway (i don't see VDDP in my HWi screens) !
You can try, run testmem5, try 1.15v, run testmem5 etc
 
Joined
Jan 5, 2006
Messages
18,347 (2.71/day)
System Name AlderLake / Laptop
Processor Intel i7 12700K P-Cores @ 5Ghz / Intel i3 7100U
Motherboard Gigabyte Z690 Aorus Master / HP 83A3 (U3E1)
Cooling Noctua NH-U12A 2 fans + Thermal Grizzly Kryonaut Extreme + 5 case fans / Fan
Memory 32GB DDR5 Corsair Dominator Platinum RGB 6000MT/s CL36 / 8GB DDR4 HyperX CL13
Video Card(s) MSI RTX 2070 Super Gaming X Trio / Intel HD620
Storage Samsung 980 Pro 1TB + 970 Evo 500GB + 850 Pro 512GB + 860 Evo 1TB x2 / Samsung 256GB M.2 SSD
Display(s) 23.8" Dell S2417DG 165Hz G-Sync 1440p / 14" 1080p IPS Glossy
Case Be quiet! Silent Base 600 - Window / HP Pavilion
Audio Device(s) Panasonic SA-PMX94 / Realtek onboard + B&O speaker system / Harman Kardon Go + Play / Logitech G533
Power Supply Seasonic Focus Plus Gold 750W / Powerbrick
Mouse Logitech MX Anywhere 2 Laser wireless / Logitech M330 wireless
Keyboard RAPOO E9270P Black 5GHz wireless / HP backlit
Software Windows 11 / Windows 10
Benchmark Scores Cinebench R23 (Single Core) 1936 @ stock Cinebench R23 (Multi Core) 23006 @ stock
Joined
Aug 9, 2019
Messages
1,567 (0.87/day)
Processor 7800X3D 2x16GB CO
Motherboard Asrock B650m HDV
Cooling Peerless Assassin SE
Memory 2x16GB DR A-die@6000c30 tuned
Video Card(s) Asus 4070 dual OC 2610@915mv
Storage WD blue 1TB nvme
Display(s) Lenovo G24-10 144Hz
Case Corsair D4000 Airflow
Power Supply EVGA GQ 650W
Software Windows 10 home 64
Benchmark Scores Superposition 8k 5267 Aida64 58.5ns
Sure. But you cannot apply X to Y here. If the board has a tendency to overshoot and supply more voltage actually, then CO numbers will be greater, you have more voltage budget. Just treat it as a knob and steps. It actually does not matter as long it performs the same.

You have to put the CPUs in the same system and force same temps to pick up a golden sample. You did it across multiple machines, thus the numbers are not comparable really.

Yeah, you can do some insight for the chatter, but it is all pretty random flavored to make some statements.

I have to agree that the tier levels is still a thing, anywhere silicon related. Even for phone CPUs and we have around 10levels. So in reality you can have same phone, but it can have 10 steps down worse voltage table, depending how lucky you are. But there you have no means to alter anything. So thank PC for being PC not Mac, and we still have the power to tinker around.
All CPUs where yested on Asrock B650m hdv ;) I forgot to mention that.

We should consider all those number only when taking account vdroop and specific motherboard configurations where this number really differs. Your 1.2V are not the same 1.2V at my board in reality during specific stress tests.

There is no way to measure the summed voltage at the core as you cannot put probes there, but those numbers should not be taken as carved in stone, they are within ~10% tolerance, and that's still a lot. Basically it is chatter about nothing really. Each specific case is on their own rules and scenarios.
I forgot to write that all cpus were tested with Asrock B650m hdv bios 2.08 :)
 
Joined
May 2, 2022
Messages
1,355 (1.69/day)
Location
G-City, UK
System Name My first new build
Processor Ryzen 7 5700X
Motherboard MSI B450M mortar max
Cooling Deepcool AK620
Memory 32gb Corsair Vengeance LPX 3200@XMP3200
Video Card(s) Sapphire 6900XT Toxic Air Cooled
Storage Samsung 870 QVO 1TB x 2, Lexar 256 GB, Western Digital SN550 500 GB nvme
Display(s) 32" LG GN600-B
Case Aerocool Eclipse
Audio Device(s) Corsair HS70 wireless
Power Supply MSI MPG AGF 850 watt gold
Mouse Glorious Model D l Pad GameSir G7 SE
Keyboard Redragon Vara K551P
Software Windows 11
Don't they sell DDR5 at the UK there?.... :D
Nope. :D It was all Amazon and ordered before I noticed the delivery date was the 16th.o_O
 
Joined
Jan 29, 2023
Messages
989 (1.86/day)
System Name KLM
Processor 7800X3D
Motherboard B-650E-E Strix
Cooling Arctic Cooling III 280
Memory 16x2 Fury Renegade 6000-32
Video Card(s) 4070-ti PNY
Storage 512+512+1+2+2+2+2+6+500+256+4+4+4
Display(s) VA 32" 4K@60 - OLED 27" 2K@240
Case 4000D Airflow
Audio Device(s) Edifier 1280Ts
Power Supply Shift 1000
Mouse 502 Hero
Keyboard K68
Software EMDB
Benchmark Scores 0>1000
You can try, run testmem5, try 1.15v, run testmem5 etc
OKay, 1.200 i'm sure, 1.5* i will try, now i'll get testmem5...

One question, since the VSoc story of x3d that burnt, what's the avantage of going from DOCP 1.25* (auto) to 1.15 (if stable) ? Oh and why auto DOCP pushed it to 1.25* ?

Thanks.
 
Last edited:
Joined
Aug 9, 2019
Messages
1,567 (0.87/day)
Processor 7800X3D 2x16GB CO
Motherboard Asrock B650m HDV
Cooling Peerless Assassin SE
Memory 2x16GB DR A-die@6000c30 tuned
Video Card(s) Asus 4070 dual OC 2610@915mv
Storage WD blue 1TB nvme
Display(s) Lenovo G24-10 144Hz
Case Corsair D4000 Airflow
Power Supply EVGA GQ 650W
Software Windows 10 home 64
Benchmark Scores Superposition 8k 5267 Aida64 58.5ns
No, try 1.15v if 1.2 passes. Load usmus1 config in testmem5.

You can probably run 6000 at 1.0-1.2v SOC. If you get down to 1.1v or lower your consumption will drop by 10W+ which will lower temps, give you a bit better allcore clocks and lower noise from cooler :)
 
Joined
Jan 29, 2023
Messages
989 (1.86/day)
System Name KLM
Processor 7800X3D
Motherboard B-650E-E Strix
Cooling Arctic Cooling III 280
Memory 16x2 Fury Renegade 6000-32
Video Card(s) 4070-ti PNY
Storage 512+512+1+2+2+2+2+6+500+256+4+4+4
Display(s) VA 32" 4K@60 - OLED 27" 2K@240
Case 4000D Airflow
Audio Device(s) Edifier 1280Ts
Power Supply Shift 1000
Mouse 502 Hero
Keyboard K68
Software EMDB
Benchmark Scores 0>1000
I'm doing 1.15, yeah i did select usmus like first post mentionned, i'll report 1.15 goes all well with testmem :p it's v0.13 git (has dll, perhaps added lastly, other i found v0.12 had not)



\\//




1.15 DONE, i go 1.1 now... 1h30min was not so long.

\\

PASSED 1.1 OKay... what do i do ?.. i'm happy and that's enough, is next step 1.05 ?

A or M die thingy ?
 
Last edited:
Joined
Nov 18, 2010
Messages
7,266 (1.46/day)
Location
Rīga, Latvia
System Name HELLSTAR
Processor AMD RYZEN 9 5950X
Motherboard ASUS Strix X570-E
Cooling 2x 360 + 280 rads. 3x Gentle Typhoons, 3x Phanteks T30, 2x TT T140 . EK-Quantum Momentum Monoblock.
Memory 4x8GB G.SKILL Trident Z RGB F4-4133C19D-16GTZR 14-16-12-30-44
Video Card(s) Sapphire Pulse RX 7900XTX. Water block. Crossflashed.
Storage Optane 900P[Fedora] + WD BLACK SN850X 4TB + 750 EVO 500GB + 1TB 980PRO+SN560 1TB(W11)
Display(s) Philips PHL BDM3270 + Acer XV242Y
Case Lian Li O11 Dynamic EVO
Audio Device(s) SMSL RAW-MDA1 DAC
Power Supply Fractal Design Newton R3 1000W
Mouse Razer Basilisk
Keyboard Razer BlackWidow V3 - Yellow Switch
Software FEDORA 40
All CPUs where yested on Asrock B650m hdv ;) I forgot to mention that.


I forgot to write that all cpus were tested with Asrock B650m hdv bios 2.08 :)

If you see variance in between CPUs why then there are not variances between boards? Those have even more variables.
 
Joined
Aug 9, 2019
Messages
1,567 (0.87/day)
Processor 7800X3D 2x16GB CO
Motherboard Asrock B650m HDV
Cooling Peerless Assassin SE
Memory 2x16GB DR A-die@6000c30 tuned
Video Card(s) Asus 4070 dual OC 2610@915mv
Storage WD blue 1TB nvme
Display(s) Lenovo G24-10 144Hz
Case Corsair D4000 Airflow
Power Supply EVGA GQ 650W
Software Windows 10 home 64
Benchmark Scores Superposition 8k 5267 Aida64 58.5ns
If you see variance in between CPUs why then there are not variances between boards? Those have even more variables.
Cpu 2 and 3 were tested on the same board with same bios. They have the largest imc variance. I set cpu 2 in an Asus B650 tuf and it needed the excact same CO, SOC, VDDP etc as the Adrock-board :)
 

tabascosauz

Moderator
Supporter
Staff member
Joined
Jun 24, 2015
Messages
7,875 (2.38/day)
Location
Western Canada
System Name ab┃ob
Processor 7800X3D┃5800X3D
Motherboard B650E PG-ITX┃X570 Impact
Cooling NH-U12A + T30┃AXP120-x67
Memory 64GB 6400CL32┃32GB 3600CL14
Video Card(s) RTX 4070 Ti Eagle┃RTX A2000
Storage 8TB of SSDs┃1TB SN550
Case Caselabs S3┃Lazer3D HT5
If you see variance in between CPUs why then there are not variances between boards? Those have even more variables.

I really don't know what you're after here. No one said boards don't have variances too, we've all known that since AM4. Just sharing a little bit of user data on 7800X3Ds.

Yes, hardware probes are best, but SVI2/SVI3 measurements are not bad at all. This is the first CPU I haven't tested in multiple boards, but the previous 5 CPUs all saw testing in 3+ boards each. There are some differences in resulting SOC power measurements between boards, but there aren't huge variations in VSOC requirements if the CPU doesn't change.

Vdroop is not that big a deal either. There will be a bit when UMC and Fabric are stressed in TM5, but we aren't dealing with the high current draw big iGPUs here. Those can really tank VSOC when they pull dozens of amps.

I'm doing 1.15, yeah i did select usmus like first post mentionned, i'll report 1.15 goes all well with testmem :p it's v0.13 git (has dll, perhaps added lastly, other i found v0.12 had not)

At the bottom end, VSOC stability is not always super clear. ie. I can get away with 1.0V just fine, but it threw a single error once hours into 1usmusv3. 1.0V gets pretty sketchy with iGPU on, but with iGPU off it's okay. So I usually run 1.05V just to not have to worry about it. Try to get it below 1.1V if you can - at that point SOC power is low enough that going even further might not be worth it.

So I've purchased a 7800X3D and 6000 CL30 trident z. Can't I just put it all together, enable expo and be fine?

I'm asking as I have everything except the ram which is coming from the USofA.

No one can promise you anything. You always have to memtest, even EXPO if you value your data. But it should be a pretty safe bet at this point in time. Fortunately it's kind of a one stop shop these days, TM5 on DDR5 is quite good at hitting two birds with one stone - RAM stability and UMC stability.
 
Last edited:
Joined
Nov 18, 2010
Messages
7,266 (1.46/day)
Location
Rīga, Latvia
System Name HELLSTAR
Processor AMD RYZEN 9 5950X
Motherboard ASUS Strix X570-E
Cooling 2x 360 + 280 rads. 3x Gentle Typhoons, 3x Phanteks T30, 2x TT T140 . EK-Quantum Momentum Monoblock.
Memory 4x8GB G.SKILL Trident Z RGB F4-4133C19D-16GTZR 14-16-12-30-44
Video Card(s) Sapphire Pulse RX 7900XTX. Water block. Crossflashed.
Storage Optane 900P[Fedora] + WD BLACK SN850X 4TB + 750 EVO 500GB + 1TB 980PRO+SN560 1TB(W11)
Display(s) Philips PHL BDM3270 + Acer XV242Y
Case Lian Li O11 Dynamic EVO
Audio Device(s) SMSL RAW-MDA1 DAC
Power Supply Fractal Design Newton R3 1000W
Mouse Razer Basilisk
Keyboard Razer BlackWidow V3 - Yellow Switch
Software FEDORA 40
I really don't know what you're after here. No one said boards don't have variances too, we've all known that since AM4. Just sharing a little bit of user data on 7800X3Ds.

We are still in the realm under 10%, witch is still high. We get different results even swapping RAM slots around as the AGESA training at boot change internal impendences.

My point remains in the fact that this is all data that should be taken with mountain load of salt. Any voltage reading without seeing the graph in recorded time frame in microseconds is softly speaking... waste of breath, it is suggestive.

We see the numbers, yes, but those don't say anything really. Too many variables to make solid deductions in each specific case even more screwing around with sloppy written code atop of it. There is no point discussing it really.
 
Joined
Jan 29, 2023
Messages
989 (1.86/day)
System Name KLM
Processor 7800X3D
Motherboard B-650E-E Strix
Cooling Arctic Cooling III 280
Memory 16x2 Fury Renegade 6000-32
Video Card(s) 4070-ti PNY
Storage 512+512+1+2+2+2+2+6+500+256+4+4+4
Display(s) VA 32" 4K@60 - OLED 27" 2K@240
Case 4000D Airflow
Audio Device(s) Edifier 1280Ts
Power Supply Shift 1000
Mouse 502 Hero
Keyboard K68
Software EMDB
Benchmark Scores 0>1000

@tabascosauz


OKay, i'll do 1.05 and call it a day...

... for RAM full test, after the 1h30m usmus should i do another different too ?



ERROR x1, after less than 1min !.. 1.07 to do now instead...
 
Last edited:
Top