Jump to content

He's stressing everything with words like "you have to" and "otherwise it won't run properly", implying there are serious consequences to which there are not.

 

All Ryzen 3rd gen does when you set voltage too low for that frequency is that the performance starts to drop even though it doesn't crash, that's why you have to validate each voltage and frequency setting with benchmarks on top of stability, say Cinebench R20 and 3DMark Fire Strike physics score. If you think stock voltages (safe btw) makes you uncomfortable, you can undervolt and underclock the CPU, no one's stopping you. 

 

Cool'n quiet is just a power saving feature, with it disabled the CPU will just burn more power.

 

None of these cause irreversible damage.

 

Bad software choice too. Since Zen 2's clock modulation works faster than other CPUs, you need Ryzen Master to read the clocks properly.

 

The only thing totally correct is that the CPU runs a single core at max speed and max voltage in the bios. AMD did make the power management on single threaded workloads super aggressive, way too aggressive if you ask me. Guess they are doing everything to break the "AMD single core performance suck" image.

CPU: i7-2600K 4751MHz 1.44V (software) --> 1.47V at the back of the socket Motherboard: Asrock Z77 Extreme4 (BCLK: 103.3MHz) CPU Cooler: Noctua NH-D15 RAM: Adata XPG 2x8GB DDR3 (XMP: 2133MHz 10-11-11-30 CR2, custom: 2203MHz 10-11-10-26 CR1 tRFC:230 tREFI:14000) GPU: Asus GTX 1070 Dual (Super Jetstream vbios, +70(2025-2088MHz)/+400(8.8Gbps)) SSD: Samsung 840 Pro 256GB (main boot drive), Transcend SSD370 128GB PSU: Seasonic X-660 80+ Gold Case: Antec P110 Silent, 5 intakes 1 exhaust Monitor: AOC G2460PF 1080p 144Hz (150Hz max w/ DP, 121Hz max w/ HDMI) TN panel Keyboard: Logitech G610 Orion (Cherry MX Blue) with SteelSeries Apex M260 keycaps Mouse: BenQ Zowie FK1

 

Model: HP Omen 17 17-an110ca CPU: i7-8750H (0.125V core & cache, 50mV SA undervolt) GPU: GTX 1060 6GB Mobile (+80/+450, 1650MHz~1750MHz 0.78V~0.85V) RAM: 8+8GB DDR4-2400 18-17-17-39 2T Storage: HP EX920 1TB PCIe x4 M.2 SSD + Crucial MX500 1TB 2.5" SATA SSD, 128GB Toshiba PCIe x2 M.2 SSD (KBG30ZMV128G) gone cooking externally, 1TB Seagate 7200RPM 2.5" HDD (ST1000LM049-2GH172) left outside Monitor: 1080p 126Hz IPS G-sync

 

Desktop benching:

Cinebench R15 Single thread:168 Multi-thread: 833 

SuperPi (v1.5 from Techpowerup, PI value output) 16K: 0.100s 1M: 8.255s 32M: 7m 45.93s

Link to post
Share on other sites

3 minutes ago, Jurrunio said:

He's stressing everything with words like "you have to" and "otherwise it won't run properly", implying there are serious consequences to which there are not.

 

All Ryzen 3rd gen does when you set voltage too low for that frequency is that the performance starts to drop even though it doesn't crash, that's why you have to validate each voltage and frequency setting with benchmarks on top of stability, say Cinebench R20 and 3DMark Fire Strike physics score. If you think stock voltages (safe btw) makes you uncomfortable, you can undervolt and underclock the CPU, no one's stopping you. 

 

Cool'n quiet is just a power saving feature, with it disabled the CPU will just burn more power.

 

None of these cause irreversible damage.

 

Bad software choice too. Since Zen 2's clock modulation works faster than other CPUs, you need Ryzen Master to read the clocks properly.

 

The only thing totally correct is that the CPU runs a single core at max speed and max voltage in the bios. AMD did make the power management on single threaded workloads super aggressive, way too aggressive if you ask me. Guess they are doing everything to break the "AMD single core performance suck" image.

 

 

Intel wish's they did that. They would bee selling 5.3ghz i9's lol.

 

But year there single core voltages are through the roof. I mean there safe and will last at a min of 3 years but PBO gets really aggressive with single core voltages. I seen 3600's pull 65w fur 100% load across all cores then pbo put it to 65w load for 1 core. Thats just crazy.

Link to post
Share on other sites

1 minute ago, bignaz said:

But year there single core voltages are through the roof. I mean there safe and will last at a min of 3 years but PBO gets really aggressive with single core voltages. I seen 3600's pull 65w fur 100% load across all cores then pbo put it to 65w load for 1 core. Thats just crazy.

I hope they dont do the same on laptops. A mere annoyance on desktop will be a massive throttling pain on laptops.

CPU: i7-2600K 4751MHz 1.44V (software) --> 1.47V at the back of the socket Motherboard: Asrock Z77 Extreme4 (BCLK: 103.3MHz) CPU Cooler: Noctua NH-D15 RAM: Adata XPG 2x8GB DDR3 (XMP: 2133MHz 10-11-11-30 CR2, custom: 2203MHz 10-11-10-26 CR1 tRFC:230 tREFI:14000) GPU: Asus GTX 1070 Dual (Super Jetstream vbios, +70(2025-2088MHz)/+400(8.8Gbps)) SSD: Samsung 840 Pro 256GB (main boot drive), Transcend SSD370 128GB PSU: Seasonic X-660 80+ Gold Case: Antec P110 Silent, 5 intakes 1 exhaust Monitor: AOC G2460PF 1080p 144Hz (150Hz max w/ DP, 121Hz max w/ HDMI) TN panel Keyboard: Logitech G610 Orion (Cherry MX Blue) with SteelSeries Apex M260 keycaps Mouse: BenQ Zowie FK1

 

Model: HP Omen 17 17-an110ca CPU: i7-8750H (0.125V core & cache, 50mV SA undervolt) GPU: GTX 1060 6GB Mobile (+80/+450, 1650MHz~1750MHz 0.78V~0.85V) RAM: 8+8GB DDR4-2400 18-17-17-39 2T Storage: HP EX920 1TB PCIe x4 M.2 SSD + Crucial MX500 1TB 2.5" SATA SSD, 128GB Toshiba PCIe x2 M.2 SSD (KBG30ZMV128G) gone cooking externally, 1TB Seagate 7200RPM 2.5" HDD (ST1000LM049-2GH172) left outside Monitor: 1080p 126Hz IPS G-sync

 

Desktop benching:

Cinebench R15 Single thread:168 Multi-thread: 833 

SuperPi (v1.5 from Techpowerup, PI value output) 16K: 0.100s 1M: 8.255s 32M: 7m 45.93s

Link to post
Share on other sites

19 hours ago, Jurrunio said:

All Ryzen 3rd gen does when you set voltage too low for that frequency is that the performance starts to drop even though it doesn't crash, that's why you have to validate each voltage and frequency setting with benchmarks on top of stability, say Cinebench R20 and 3DMark Fire Strike physics score. If you think stock voltages (safe btw) makes you uncomfortable, you can undervolt and underclock the CPU, no one's stopping you. 

Im not worried about stock really but since anthony recommended a 1,35 V core i was wondering if that does actually reduce single core performance or alternatively is a big waste of electricity of they cant tune down the voltage on lower loads

Link to post
Share on other sites

12 minutes ago, Eumel2 said:

Im not worried about stock really but since anthony recommended a 1,35 V core i was wondering if that does actually reduce single core performance or alternatively is a big waste of electricity of they cant tune down the voltage on lower loads

It may or may not drop performance. It depends on the CPU you're using (silicon lottery). Anthony presumably is being safe here since it's risky to recommend aggressive voltages without knowing whether something else nad is going on

CPU: i7-2600K 4751MHz 1.44V (software) --> 1.47V at the back of the socket Motherboard: Asrock Z77 Extreme4 (BCLK: 103.3MHz) CPU Cooler: Noctua NH-D15 RAM: Adata XPG 2x8GB DDR3 (XMP: 2133MHz 10-11-11-30 CR2, custom: 2203MHz 10-11-10-26 CR1 tRFC:230 tREFI:14000) GPU: Asus GTX 1070 Dual (Super Jetstream vbios, +70(2025-2088MHz)/+400(8.8Gbps)) SSD: Samsung 840 Pro 256GB (main boot drive), Transcend SSD370 128GB PSU: Seasonic X-660 80+ Gold Case: Antec P110 Silent, 5 intakes 1 exhaust Monitor: AOC G2460PF 1080p 144Hz (150Hz max w/ DP, 121Hz max w/ HDMI) TN panel Keyboard: Logitech G610 Orion (Cherry MX Blue) with SteelSeries Apex M260 keycaps Mouse: BenQ Zowie FK1

 

Model: HP Omen 17 17-an110ca CPU: i7-8750H (0.125V core & cache, 50mV SA undervolt) GPU: GTX 1060 6GB Mobile (+80/+450, 1650MHz~1750MHz 0.78V~0.85V) RAM: 8+8GB DDR4-2400 18-17-17-39 2T Storage: HP EX920 1TB PCIe x4 M.2 SSD + Crucial MX500 1TB 2.5" SATA SSD, 128GB Toshiba PCIe x2 M.2 SSD (KBG30ZMV128G) gone cooking externally, 1TB Seagate 7200RPM 2.5" HDD (ST1000LM049-2GH172) left outside Monitor: 1080p 126Hz IPS G-sync

 

Desktop benching:

Cinebench R15 Single thread:168 Multi-thread: 833 

SuperPi (v1.5 from Techpowerup, PI value output) 16K: 0.100s 1M: 8.255s 32M: 7m 45.93s

Link to post
Share on other sites

On 8/27/2019 at 5:02 AM, Jurrunio said:

It may or may not drop performance. It depends on the CPU you're using (silicon lottery). Anthony presumably is being safe here since it's risky to recommend aggressive voltages without knowing whether something else nad is going on

Basically yeah. I was also taken to task on it as being too high in the overclocking guide video as a result of The Stilt's findings on silicon degradation past 1.325V... But he's talking about high currents. In other words, setting even higher than 1.35V isn't going to hit 1.325V at full load due to vdroop unless you've got really aggressive LLC enabled, so it's a safe value.

 

Steve over at Gamers Nexus did a video on undervolting Ryzen 3rd-gen and found that Ryzen can and will undervolt well, unless you go too low. In the case of the overclocks I was able to achieve, thermals were the big constraining factor on the 3900X, while the 3700X and 3600 were limited primarily in terms of silicon - Higher voltages did nothing for stability nor performance.

 

TL;DR 1.35V maximum manual voltage setting is mainly a guideline based on my observations and not wanting to stray too close to the sun on silicon degradation. The maximum safe voltage in general was determined to be 1.47V, although transient voltages will be higher than that even at stock and that's fine - It's sustained you need to worry about.

 

A note regarding voltages and hardware monitors: Some core voltages will remain "high" at idle as a result of software being incapable of reading the state of a parked core, which means the last measurement before parking becomes the "current" measurement whether it's accurate or not. Ryzen Master can show the status of each core and their true current clocks, but unfortunately doesn't show each core's voltage as HWINFO does.

 

Much of this is also going to vary from board to board, BIOS to BIOS, and chip to chip. Early BIOSes had poor voltage management, so in these cases the CPUs might have been starving or otherwise not boosting properly. This is less of a concern now.

Link to post
Share on other sites

  • 5 months later...

Create an account or sign in to comment

You need to be a member in order to leave a comment

Create an account

Sign up for a new account in our community. It's easy!

Register a new account

Sign in

Already have an account? Sign in here.

Sign In Now

×