*Official Benchmark Thread* - Post it here or it didn't happen :D

[jaybee83]

8 hours ago, electrosoft said:

 

LOL, if you were "safe" to upgrade from your 2080ti to a 3090, you are definitely safe to upgrade from a 3090 to a 4090. 🤣

 

I hope the 8900xtx smashes Nvidia into the ground because the 7900xtx was a bit of a let down after all the pre-launch hype and pseudo CU count that AMD didn't correct properly and let speculation fly amok.

 

I said it before and I'll say it again, you will get a 4080ti dab smack in the middle of the 4080 and 4090 for ~1299 or 1399 with 20GB and a slower memory bus.

 

I am guessing we might *might* get a 4090ti Summer of 2024 and the 5090 drops Q1 2025. Either scenario, if you have a 4090, a 4090ti is DOA as 5090 is right around the corner. Don't pull an "Electrosoft" 🤣 and get all nostalgic and order a KPE 3090ti for $2499 and end up regretting it (But then EVGA refunded me the $500 difference when it dropped and a collector wanted it for $2k so I broke even and am still using the included 1600w P2 to this day but still....)

 

You can always sit back and wait and compromise or you just order the new king launch card, ride it out for the entire cycle, get the new king launch card, sell your old card (or not) for what you can get...rinse/repeat.

 

My only problem when the 5090 drops is Fallout76 and WoW have met their 4k Ultra match even with RT enabled with the 4090 for 144hz displays. For the first time, I am hard pressed to find a compelling reason to upgrade next cycle outside of the ever present, "I want it."

 

I am rapidly coming up on a full year of owning this MSI  4090 Liquid Suprim X in one month and it has been an absolute joy and beast of a card. Outside of the price, I have zero complaints.

 

 

 

 

 

 

 

couldnt agree more with this. the main factor i focus on is "usage time". so get the king card right when its released and enjoy it for the longest time. IF a 4090 Ti ever comes out (which i doubt, no monetary incentive for Nvidia), ill be totally chill without any need to upgrade. at best, with full fat 102 die, itll be what? maybe 10-15% more performance? plus with my card oced to the max that gap gets quite small, so couldnt be bothered to upgrade. as you said, ill just wait and see what the 5090 brings to the table, until then i can save up enough money to do that upgrade without having to sell off my 4090 beforehand. no downtime, max usage time and some money back from selling off my 4090 in the end 🙂 and if the 5090 is NOT available ANYwhere due to "insert random tech hype for scalper prices" then i still have my 4090, all good!

 

5 hours ago, Papusan said:

Yup, the Arctic Liquid Freezer II is still a good AIO. Even better if you swap the fans to something more powerful. No need to buy AIOs with all the disgusting bling bling all over the places

 

How good is cheap? Arctic Liquid Freezer II 240 and 360 in the test – This standard has to be beaten igorslab

 

 

A new 4080Ti won't be sold in China. A real scalpers card. Cheap enough to make proper profits when you ship it over to China. Hmmm. I wonder how high the Global price hike will be above MSRP. 

 

oh yes, very happy with my AC LF II 420, especially with the 6x Noctua IndustriallPPC 3000 fans in push-pull, quite the cooling beast 😄 

[Mr. Fox]

21 hours ago, Papusan said:

Why would you create new xx80 GPU's that can't be sold world wide ? And why would you use defective AD102 dies that instead can go directly into the more expensive 4090D/workstation cards for the Chinese market? Hmm... Food for thoughts.

It looks like they are going to scrap the 4090D now, so that's good. A world without China's participation would be a better place for everyone else.  It does not really matter what the law says, China will steal whatever they want if they can't buy it. They do that already even when they can buy it. Attempting to skirt a law designed to hurt or hinder China's success and prosperity should carry a severe punishment by itself for NVIDIA or any other company looking for a loophole. It should be viewed as a form of treason and aiding and abetting organized crime. In addition to that, payback for COVID should be severe and have no expiration date. Those involved on our side (Fauchi and company) should be looking at facing capital punishment. Ask a Uyghur that has had family members executed or a Christian that has been imprisoned or executed for their faith how they feel about China's government.

[Papusan]

16 hours ago, Mr. Fox said:

The only way this can ever get fixed is if enough people wise up and make it impossible for the imbeciles scalping the public to earn a living by forcing them into bankruptcy and shutting down their businesses with aging inventory that is only sellable at a price less than what they paid for it. I doubt there are enough intelligent people for that to happen, but I would love to be proven wrong.

 

Be you sure... The world is filled with people (consumers or shall we say stupids) that will jump on everything. That's also the reason Nvidia can continue screw the consumers year after year....

 

What to do if your new and flawed doesn't sell.... Destroy your older gen cards to try push consumers over on their new overpriced trash. 

Even weaker Nvidia RTX 3050 6GB reportedly to replace the 3050 8GB

 

Yup, no more a graphics card company. Now only a greedy Unisex company that value profits over offering the best for the consumers for an ok'ish price... 

 

Nvidia is ‘no longer a graphics company’ digitaltrends.com

 

9 hours ago, Mr. Fox said:

It looks like they are going to scrap the 4090D now, so that's good. A world without China's participation would be a better place for everyone else.  It does not really matter what the law says, China will steal whatever they want if they can't buy it.

 

Imagine if Nvidia could continue sell all 4090 cards to China. The mining boom is nothing compared to what to come from AI and China. A big business we have never seen before, would pop up transforming loads if not all of 4090's (grabbed from all the worlds corners Nvidia AIC partners) into AI graphics accelerators for China. None would find 4090's at normal prices anymore. All cards would go into China. 

[jaybee83]

Aight gaiz, i know its been LONG overdue now, but finally getting around to writing up my AM5 / Ryzen 7000 DDR5 OC adventures 😄 hope this helps my fellow AM5 owners!

 

Stress testing: Ive been using Tm5 with Anta777 Absolut profile, which gave me very nice results in terms of stability. anything that passes that will pass everything else in my everyday usage. For my current 2x24GB kit, it needs roughly 3 hours to complete at SPD settings (5200 Mhz) and 2 hours at tuned settings (6200 or 7600 with tightened timings). I have also seen lots of recommendations for Karhu, but havent tested that yet.

 

Infinity Fabric (IF / FCLK)

- Stock is at 1800 - 2000 Mhz, depending on motherboard and bios revision

- Depending on CPU quality, typically maxes out around 2133 to 2200 Mhz (in my case 2167 Mhz)

- As opposed to Ryzen 5000, the IF is now decoupled from the Memory Controller (MCLK) and Memory itself (UCLK), so you can run it asynchronously

- Depending on the setup and settings, I found that sometimes synchronous, other times maxxed out IF clocks are better for the resulting memory latency (i.e. DDR5-6200 with IF at 2067 = 3x ratio synchronous vs. DDR5-6200 with IF 2167 = 2.86 asynchronous), so best to check both!

 

"Gears" = UCLK/MCLK ratios

- "Gear 1" = UCLK/MCLK = 1:1 ratio, i.e. Memory Controller at 3000 Mhz and DDR5-6000 at 3000 Mhz actual clocks

- "Gear 2" = UCLK/MCLK = 1:2 ratio, i.e.  i.e. Memory Controller at 2000 Mhz and DDR5-8000 at 4000 Mhz actual clocks

- As with Intel, Gear 2 has a latency penalty but allows for much higher RAM frequencies and thus bandwidth.

- In my case, I maxed out at the following:

                      DDR5-6400 bootable and DDR5-6200 stable in Gear 1

                      DDR5-8000 bootable and DDR5-7600 stable in Gear 2

- From what ive seen, typical max ranges for Gear 1 are 6200-6600, for Gear 2 7600-8200. How stable those were though, I cannot judge 🙂 your results WILL vary depending on motherboard, CPU IMC quality and RAM sticks, as per usual...

 

Voltages

- VDDG does not have any impact on RAM OC (at least in my case, see next point directly below)

- The Misc voltage is a general power delivery path from the mobo to the CPU, the VDDG voltage is actually derived from that and controls the IF / FCLK voltage. funny enough, i had to increase the Misc voltage for RAM OC but VDDG didnt have any impact whatsoever, either for IF or RAM OC. I think that the stock VDDG voltage is already enough for any kind of IF or RAM OC, at least in my case. Could be that at even higher RAM or IF speeds one would need to also adjust VDDG in the future.

- SOC, VDDIO, Misc, VDDP, as well as RAM VDD and RAM VDDQ are important for RAM OC on AM5

- With a bit of trial and error, as well as lots of researching and comparisons I have found the following voltage ranges safe for testing (all of the listed max values I have been running for months during tuning of timings, so at least from my side I consider those safe for 24/7, your mileage may vary!):

SOC = Stock 1.00V, max. 1.30V (max possible with newer Bios revisions)

VDDIO = Stock 1.10V, max. 1.58V

Misc = Stock 1.10V, max. 1.55 V

VDDP = Stock 0.80V, max. 1.18V

RAM VDD = Stock 1.10V, max. 1.63V

RAM VDDQ = Stock 1.10V, max. 1.63V

 

Naturally, after finding ur best settings for timings, I would suggest to go back and lower the above voltages as much as possible to save on voltage wear and heat as much as possible. Here were the lowest settings I found to be stable at my tightened timings:

 

DDR5-6200 Gear 1 (G.Skill 2x16GB Kit DDR5-6600)

SOC = 1.20V

VDDIO = 1.31V

Misc = 1.10V (Stock)

VDDP = 1.03V

RAM VDD = 1.52V

RAM VDDQ = 1.52V

 

DDR5-6200 Gear 1 (Teamgroup 2x24GB Kit DDR5-8200)

SOC = 1.19V

VDDIO = 1.27V

Misc = 1.39V

VDDP = 0.81V

RAM VDD = 1.58V

RAM VDDQ = 1.59V

 

DDR5-7600 Gear 2 (Teamgroup 2x24GB Kit DDR5-8200)

SOC = 1.26V

VDDIO = 1.52V

Misc = 1.26V

VDDP = 1.10V

RAM VDD = 1.59V

RAM VDDQ = 1.59V

 

Notes:

- VDD can sometimes be lowered further than VDDQ, so check and see whats possible here. In my case only 0.01V was possible.

- As you can see, minimum required voltages are also different for each ram kit, so your results may vary, even with the same setup otherwise.

 

Additional Settings

Nitro Mode

The most recent Bios updates for AM5 boards, a new "Nitro Mode" has been implemented, which allows a more finegrained tuning of RAM training. Loosening up these settings allows for more stability at the cost of latency and potentially longer boot times. It includes the following settings:

 

Nitro Mode = Enabled / Disabled

Robust Training = Enabled / Disabled (Enable for better stability)

Nitro RX Data / Nitro Tx Data / Nitro Control Line / Nitro Rx Burst Length / Nitro Tx Burst Length = basically, the higher the factor, the higher the stability / latency / boot time.

 

What I did was to max out everything, same with the voltages, then tighten up all the timings, afterwards go back and find the lowest values that are still stable to improve on latency and boot times.

 

Other

There are two additional settings important for RAM tuning in this case: "Memory Context Restore" and "Power Down Enable"

For best latency and performance (i.e. reaching tighter timings / higher clocks) I recommend disabling them both. This will, however, significantly increase our boot times due to RAM retraining going on for every boot.

I leave them both disabled cuz i dont find the longer boot times that bad, were talking maybe 30-40 seconds vs. 10 seconds, definitely worth the tradeoff for better (and stable!) performance 🙂 

 

Cooling

This is more of a general thing and not specific to AM5: DDR5 RAM is finicky when it comes to temperatures! In my case I leave the RGB switched off, which bring about 5-10C lower temps (crazy, right?). In addition, I have a 140mm fan pointed directly at the RAM sticks to further lower the temps. I am usually around 50C under full load with TM5 stress testing and in the high 40s when gaming and the whole case is heated up by the GPU and CPU. Ideally, you would replace the stock heatsinks with aftermarket ones or add liquid cooling to it.

Whatever you do, just keep an eye on temps! The lower, the more tight the timings and higher the clocks can be with good stability 🙂 

 

Timings

I am including my tuned timing results for both my previous G.Skill 2x16GB DDR5-6600, as well as my current Teamgroup 2x24GB DDR-8200 kits.

 

 

Notes:

- The G.Skill kit SPD was at DDR5-4800, whereas the Teamgroup kit sports DDR5-5200 at stock.

- The G.Skill kit (on an older bios) profited from synchronous IF, whereas the Teamgroup kit preferred the max. IF clock at 2167 Mhz in both Gear 1 and Gear 2.

- For tightening the timings, i maxed out the voltages as well as Nitro Mode settings and then went back after tuning to lower everything again as much as possible.

 

Results

I checked performance with AIDA64, 10 consecutive runs for each setup and calculated the average for each parameter to account for any variability. I am using the DDR5-4800 G.Skill stock as the 100% baseline here.

 

G.Skill Stock DDR5-4800

Read = 54835

Write = 52866

Copy = 52664

Latency = 96.1

 

G.Skill Tuned DDR5-6200 Gear 1

Read = 90968 (+66%)

Write = 94858 (+79%)

Copy = 81608 (+55%)

Latency = 55.4 (+74%)

 

Teamgroup Stock DDR5-5200

Read = 64214 (+17%)

Write = 64992 (+23%)

Copy = 58463 (+11%)

Latency = 94.7 (+1%)

 

Teamgroup Tuned DDR5-6200 Gear 1

Read = 91237 (+66%)

Write = 95211 (+80%)

Copy = 85265 (+62%)

Latency = 54.6 (+76%)

 

Teamgroup Tuned DDR5-7600 Gear 2

Read = 99155 (+81%)

Write = 101802 (+93%)

Copy = 91402 (+74%)

Latency = 57.9 (+66%)

 

Conclusions

- Based on the obtained results and input that i saw from other AM5 users online, I am likely held back by the CPU IMC. The RAM sticks did not show much difference between the two kits in Gear 1. Also, I have seen differing results on the X670E Extreme and Hero boards, so that leads me to think that the CPU is my current bottleneck. Curious to see what Zen 5 will bring!

-  As mentioned before, Gear 2 lets the latency suffer a bit, but provides a nice boost to bandwidth. I will stick to my Gear 2 settings for now, as the latency penalty is a mere 3 ns in my case, so nothing to worry about.

 

Overall quite happy with the results at roughly 60-90% performance uplift vs. stock speeds of my previous G.Skill RAM kit 🙂I didnt even really bother testing with the XMP / EXPO profiles tbh, just used them as a quick baseline for additional tuning 😄

 

Aight, with GPU and RAM tuning out of the way, Ill now shift my focus on tweaking my 7950X 🙂 

[Mr. Fox]

3 hours ago, jaybee83 said:

Aight gaiz, i know its been LONG overdue now, but finally getting around to writing up my AM5 / Ryzen 7000 DDR5 OC adventures 😄 hope this helps my fellow AM5 owners!

 

Stress testing: Ive been using Tm5 with Anta777 Absolut profile, which gave me very nice results in terms of stability. anything that passes that will pass everything else in my everyday usage. For my current 2x24GB kit, it needs roughly 3 hours to complete at SPD settings (5200 Mhz) and 2 hours at tuned settings (6200 or 7600 with tightened timings). I have also seen lots of recommendations for Karhu, but havent tested that yet.

 

Infinity Fabric (IF / FCLK)

- Stock is at 1800 - 2000 Mhz, depending on motherboard and bios revision

- Depending on CPU quality, typically maxes out around 2133 to 2200 Mhz (in my case 2167 Mhz)

- As opposed to Ryzen 5000, the IF is now decoupled from the Memory Controller (MCLK) and Memory itself (UCLK), so you can run it asynchronously

- Depending on the setup and settings, I found that sometimes synchronous, other times maxxed out IF clocks are better for the resulting memory latency (i.e. DDR5-6200 with IF at 2067 = 3x ratio synchronous vs. DDR5-6200 with IF 2167 = 2.86 asynchronous), so best to check both!

 

"Gears" = UCLK/MCLK ratios

- "Gear 1" = UCLK/MCLK = 1:1 ratio, i.e. Memory Controller at 3000 Mhz and DDR5-6000 at 3000 Mhz actual clocks

- "Gear 2" = UCLK/MCLK = 1:2 ratio, i.e.  i.e. Memory Controller at 2000 Mhz and DDR5-8000 at 4000 Mhz actual clocks

- As with Intel, Gear 2 has a latency penalty but allows for much higher RAM frequencies and thus bandwidth.

- In my case, I maxed out at the following:

                      DDR5-6400 bootable and DDR5-6200 stable in Gear 1

                      DDR5-8000 bootable and DDR5-7600 stable in Gear 2

- From what ive seen, typical max ranges for Gear 1 are 6200-6600, for Gear 2 7600-8200. How stable those were though, I cannot judge 🙂 your results WILL vary depending on motherboard, CPU IMC quality and RAM sticks, as per usual...

 

Voltages

- VDDG does not have any impact on RAM OC

- SOC, VDDIO, Misc, VDDP, as well as RAM VDD and RAM VDDQ are important for RAM OC on AM5

- With a bit of trial and error, as well as lots of researching and comparisons I have found the following voltage ranges safe for testing (all of the listed max values I have been running for months during tuning of timings, so at least from my side I consider those safe for 24/7, your mileage may vary!):

SOC = Stock 1.00V, max. 1.30V (max possible with newer Bios revisions)

VDDIO = Stock 1.10V, max. 1.58V

Misc = Stock 1.10V, max. 1.55 V

VDDP = Stock 0.80V, max. 1.18V

RAM VDD = Stock 1.10V, max. 1.63V

RAM VDDQ = Stock 1.10V, max. 1.63V

 

Naturally, after finding ur best settings for timings, I would suggest to go back and lower the above voltages as much as possible to save on voltage wear and heat as much as possible. Here were the lowest settings I found to be stable at my tightened timings:

 

DDR5-6200 Gear 1 (G.Skill 2x16GB Kit DDR5-6600)

SOC = 1.20V

VDDIO = 1.31V

Misc = 1.10V (Stock)

VDDP = 1.03V

RAM VDD = 1.52V

RAM VDDQ = 1.52V

 

DDR5-6200 Gear 1 (Teamgroup 2x24GB Kit DDR5-8200)

SOC = 1.19V

VDDIO = 1.27V

Misc = 1.39V

VDDP = 0.81V

RAM VDD = 1.58V

RAM VDDQ = 1.59V

 

DDR5-7600 Gear 2 (Teamgroup 2x24GB Kit DDR5-8200)

SOC = 1.26V

VDDIO = 1.52V

Misc = 1.26V

VDDP = 1.10V

RAM VDD = 1.59V

RAM VDDQ = 1.59V

 

Notes:

- VDD can sometimes be lowered further than VDDQ, so check and see whats possible here. In my case only 0.01V was possible.

- As you can see, minimum required voltages are also different for each ram kit, so your results may vary, even with the same setup otherwise.

 

Additional Settings

Nitro Mode

The most recent Bios updates for AM5 boards, a new "Nitro Mode" has been implemented, which allows a more finegrained tuning of RAM training. Loosening up these settings allows for more stability at the cost of latency and potentially longer boot times. It includes the following settings:

 

Nitro Mode = Enabled / Disabled

Robust Training = Enabled / Disabled (Enable for better stability)

Nitro RX Data / Nitro Tx Data / Nitro Control Line / Nitro Rx Burst Length / Nitro Tx Burst Length = basically, the higher the factor, the higher the stability / latency / boot time.

 

What I did was to max out everything, same with the voltages, then tighten up all the timings, afterwards go back and find the lowest values that are still stable to improve on latency and boot times.

 

Other

There are two additional settings important for RAM tuning in this case: "Memory Context Restore" and "Power Down Enable"

For best latency and performance (i.e. reaching tighter timings / higher clocks) I recommend disabling them both. This will, however, significantly increase our boot times due to RAM retraining going on for every boot.

I leave them both disabled cuz i dont find the longer boot times that bad, were talking maybe 30-40 seconds vs. 10 seconds, definitely worth the tradeoff for better (and stable!) performance 🙂 

 

Cooling

This is more of a general thing and not specific to AM5: DDR5 RAM is finicky when it comes to temperatures! In my case I leave the RGB switched off, which bring about 5-10C lower temps (crazy, right?). In addition, I have a 140mm fan pointed directly at the RAM sticks to further lower the temps. I am usually around 50C under full load with TM5 stress testing and in the high 40s when gaming and the whole case is heated up by the GPU and CPU. Ideally, you would replace the stock heatsinks with aftermarket ones or add liquid cooling to it.

Whatever you do, just keep an eye on temps! The lower, the more tight the timings and higher the clocks can be with good stability 🙂 

 

Timings

I am including my tuned timing results for both my previous G.Skill 2x16GB DDR5-6600, as well as my current Teamgroup 2x24GB DDR-8200 kits.

 

 

Notes:

- The G.Skill kit SPD was at DDR5-4800, whereas the Teamgroup kit sports DDR5-5200 at stock.

- The G.Skill kit (on an older bios) profited from synchronous IF, whereas the Teamgroup kit preferred the max. IF clock at 2167 Mhz in both Gear 1 and Gear 2.

- For tightening the timings, i maxed out the voltages as well as Nitro Mode settings and then went back after tuning to lower everything again as much as possible.

 

Results

I checked performance with AIDA64, 10 consecutive runs for each setup and calculated the average for each parameter to account for any variability. I am using the DDR5-4800 G.Skill stock as the 100% baseline here.

 

G.Skill Stock DDR5-4800

Read = 54835

Write = 52866

Copy = 52664

Latency = 96.1

 

G.Skill Tuned DDR5-6200 Gear 1

Read = 90968 (+66%)

Write = 94858 (+79%)

Copy = 81608 (+55%)

Latency = 55.4 (+74%)

 

Teamgroup Stock DDR5-5200

Read = 64214 (+17%)

Write = 64992 (+23%)

Copy = 58463 (+11%)

Latency = 94.7 (+1%)

 

Teamgroup Tuned DDR5-6200 Gear 1

Read = 91237 (+66%)

Write = 95211 (+80%)

Copy = 85265 (+62%)

Latency = 54.6 (+76%)

 

Teamgroup Tuned DDR5-7600 Gear 2

Read = 99155 (+81%)

Write = 101802 (+93%)

Copy = 91402 (+74%)

Latency = 57.9 (+66%)

 

Conclusions

- Based on the obtained results and input that i saw from other AM5 users online, I am likely held back by the CPU IMC. The RAM sticks did not show much difference between the two kits in Gear 1. Also, I have seen differing results on the X670E Extreme and Hero boards, so that leads me to think that the CPU is my current bottleneck. Curious to see what Zen 5 will bring!

-  As mentioned before, Gear 2 lets the latency suffer a bit, but provides a nice boost to bandwidth. I will stick to my Gear 2 settings for now, as the latency penalty is a mere 3 ns in my case, so nothing to worry about.

 

Overall quite happy with the results at roughly 60-90% performance uplift vs. stock speeds of my previous G.Skill RAM kit 🙂I didnt even really bother testing with the XMP / EXPO profiles tbh, just used them as a quick baseline for additional tuning 😄

 

Aight, with GPU and RAM tuning out of the way, Ill now shift my focus on tweaking my 7950X 🙂 

Nice write-up, brother.

 

Any idea what "misc" voltage is actually for?

[jaybee83]

26 minutes ago, Mr. Fox said:

Nice write-up, brother.

 

Any idea what "misc" voltage is actually for?

 

thanks, appreciate it 😀

 

very good question, actually forgot to add an info relating to that:

 

The Misc voltage is a general power delivery path from the mobo to the CPU, the VDDG voltage is actually derived from that and controls the IF / FCLK voltage. funny enough, i had to increase the Misc voltage for RAM OC but VDDG didnt have any impact whatsoever, either for IF or RAM OC. I think that the stock VDDG voltage is already enough for any kind of IF or RAM OC, at least in my case. Could be that at even higher RAM or IF speeds one would need to also adjust VDDG in the future.

 

Ill add it to the above post 🙂 

[Papusan]

On 12/6/2023 at 8:48 AM, cylix said:

Intel has lost it  🤣

 

You sure? Maybe Intel have paid Microsoft to screw up for AMD?😀

 

AMD is their own enemy but Microsoft doesn't make it easier for the consumers in the Red corner. I wonder why AMD does this to themself. Time after time AMD go into Microsofts trap. Something is wrong with the coperation between AMD and Microsoft. Could it be Intel and their deep pockets, LOOL

 

Windows 11 23H2 Is Slowing Game Performance, but Microsoft Has a Workaround extremetech.com

 

The issue seems to affect folks with AMD CPUs and results in degraded CPU performance.

 

For now, the issue seems to primarily affect AMD CPUs, including both older models and newer chips, including the Ryzen 7000-series. It's unclear what the specific problem is, but it's conceivable that AMD will release a microcode update to resolve this situation soon. If you're experiencing degraded CPU performance in games or apps, apply the workaround listed above and say a silent prayer to the silicon deities. If you have an AMD CPU and have not upgraded yet, hold off until the coast is clear.

[cylix]

 

[tps3443]

5 hours ago, jaybee83 said:

Aight gaiz, i know its been LONG overdue now, but finally getting around to writing up my AM5 / Ryzen 7000 DDR5 OC adventures 😄 hope this helps my fellow AM5 owners!

 

Stress testing: Ive been using Tm5 with Anta777 Absolut profile, which gave me very nice results in terms of stability. anything that passes that will pass everything else in my everyday usage. For my current 2x24GB kit, it needs roughly 3 hours to complete at SPD settings (5200 Mhz) and 2 hours at tuned settings (6200 or 7600 with tightened timings). I have also seen lots of recommendations for Karhu, but havent tested that yet.

 

Infinity Fabric (IF / FCLK)

- Stock is at 1800 - 2000 Mhz, depending on motherboard and bios revision

- Depending on CPU quality, typically maxes out around 2133 to 2200 Mhz (in my case 2167 Mhz)

- As opposed to Ryzen 5000, the IF is now decoupled from the Memory Controller (MCLK) and Memory itself (UCLK), so you can run it asynchronously

- Depending on the setup and settings, I found that sometimes synchronous, other times maxxed out IF clocks are better for the resulting memory latency (i.e. DDR5-6200 with IF at 2067 = 3x ratio synchronous vs. DDR5-6200 with IF 2167 = 2.86 asynchronous), so best to check both!

 

"Gears" = UCLK/MCLK ratios

- "Gear 1" = UCLK/MCLK = 1:1 ratio, i.e. Memory Controller at 3000 Mhz and DDR5-6000 at 3000 Mhz actual clocks

- "Gear 2" = UCLK/MCLK = 1:2 ratio, i.e.  i.e. Memory Controller at 2000 Mhz and DDR5-8000 at 4000 Mhz actual clocks

- As with Intel, Gear 2 has a latency penalty but allows for much higher RAM frequencies and thus bandwidth.

- In my case, I maxed out at the following:

                      DDR5-6400 bootable and DDR5-6200 stable in Gear 1

                      DDR5-8000 bootable and DDR5-7600 stable in Gear 2

- From what ive seen, typical max ranges for Gear 1 are 6200-6600, for Gear 2 7600-8200. How stable those were though, I cannot judge 🙂 your results WILL vary depending on motherboard, CPU IMC quality and RAM sticks, as per usual...

 

Voltages

- VDDG does not have any impact on RAM OC (at least in my case, see next point directly below)

- The Misc voltage is a general power delivery path from the mobo to the CPU, the VDDG voltage is actually derived from that and controls the IF / FCLK voltage. funny enough, i had to increase the Misc voltage for RAM OC but VDDG didnt have any impact whatsoever, either for IF or RAM OC. I think that the stock VDDG voltage is already enough for any kind of IF or RAM OC, at least in my case. Could be that at even higher RAM or IF speeds one would need to also adjust VDDG in the future.

- SOC, VDDIO, Misc, VDDP, as well as RAM VDD and RAM VDDQ are important for RAM OC on AM5

- With a bit of trial and error, as well as lots of researching and comparisons I have found the following voltage ranges safe for testing (all of the listed max values I have been running for months during tuning of timings, so at least from my side I consider those safe for 24/7, your mileage may vary!):

SOC = Stock 1.00V, max. 1.30V (max possible with newer Bios revisions)

VDDIO = Stock 1.10V, max. 1.58V

Misc = Stock 1.10V, max. 1.55 V

VDDP = Stock 0.80V, max. 1.18V

RAM VDD = Stock 1.10V, max. 1.63V

RAM VDDQ = Stock 1.10V, max. 1.63V

 

Naturally, after finding ur best settings for timings, I would suggest to go back and lower the above voltages as much as possible to save on voltage wear and heat as much as possible. Here were the lowest settings I found to be stable at my tightened timings:

 

DDR5-6200 Gear 1 (G.Skill 2x16GB Kit DDR5-6600)

SOC = 1.20V

VDDIO = 1.31V

Misc = 1.10V (Stock)

VDDP = 1.03V

RAM VDD = 1.52V

RAM VDDQ = 1.52V

 

DDR5-6200 Gear 1 (Teamgroup 2x24GB Kit DDR5-8200)

SOC = 1.19V

VDDIO = 1.27V

Misc = 1.39V

VDDP = 0.81V

RAM VDD = 1.58V

RAM VDDQ = 1.59V

 

DDR5-7600 Gear 2 (Teamgroup 2x24GB Kit DDR5-8200)

SOC = 1.26V

VDDIO = 1.52V

Misc = 1.26V

VDDP = 1.10V

RAM VDD = 1.59V

RAM VDDQ = 1.59V

 

Notes:

- VDD can sometimes be lowered further than VDDQ, so check and see whats possible here. In my case only 0.01V was possible.

- As you can see, minimum required voltages are also different for each ram kit, so your results may vary, even with the same setup otherwise.

 

Additional Settings

Nitro Mode

The most recent Bios updates for AM5 boards, a new "Nitro Mode" has been implemented, which allows a more finegrained tuning of RAM training. Loosening up these settings allows for more stability at the cost of latency and potentially longer boot times. It includes the following settings:

 

Nitro Mode = Enabled / Disabled

Robust Training = Enabled / Disabled (Enable for better stability)

Nitro RX Data / Nitro Tx Data / Nitro Control Line / Nitro Rx Burst Length / Nitro Tx Burst Length = basically, the higher the factor, the higher the stability / latency / boot time.

 

What I did was to max out everything, same with the voltages, then tighten up all the timings, afterwards go back and find the lowest values that are still stable to improve on latency and boot times.

 

Other

There are two additional settings important for RAM tuning in this case: "Memory Context Restore" and "Power Down Enable"

For best latency and performance (i.e. reaching tighter timings / higher clocks) I recommend disabling them both. This will, however, significantly increase our boot times due to RAM retraining going on for every boot.

I leave them both disabled cuz i dont find the longer boot times that bad, were talking maybe 30-40 seconds vs. 10 seconds, definitely worth the tradeoff for better (and stable!) performance 🙂 

 

Cooling

This is more of a general thing and not specific to AM5: DDR5 RAM is finicky when it comes to temperatures! In my case I leave the RGB switched off, which bring about 5-10C lower temps (crazy, right?). In addition, I have a 140mm fan pointed directly at the RAM sticks to further lower the temps. I am usually around 50C under full load with TM5 stress testing and in the high 40s when gaming and the whole case is heated up by the GPU and CPU. Ideally, you would replace the stock heatsinks with aftermarket ones or add liquid cooling to it.

Whatever you do, just keep an eye on temps! The lower, the more tight the timings and higher the clocks can be with good stability 🙂 

 

Timings

I am including my tuned timing results for both my previous G.Skill 2x16GB DDR5-6600, as well as my current Teamgroup 2x24GB DDR-8200 kits.

 

 

Notes:

- The G.Skill kit SPD was at DDR5-4800, whereas the Teamgroup kit sports DDR5-5200 at stock.

- The G.Skill kit (on an older bios) profited from synchronous IF, whereas the Teamgroup kit preferred the max. IF clock at 2167 Mhz in both Gear 1 and Gear 2.

- For tightening the timings, i maxed out the voltages as well as Nitro Mode settings and then went back after tuning to lower everything again as much as possible.

 

Results

I checked performance with AIDA64, 10 consecutive runs for each setup and calculated the average for each parameter to account for any variability. I am using the DDR5-4800 G.Skill stock as the 100% baseline here.

 

G.Skill Stock DDR5-4800

Read = 54835

Write = 52866

Copy = 52664

Latency = 96.1

 

G.Skill Tuned DDR5-6200 Gear 1

Read = 90968 (+66%)

Write = 94858 (+79%)

Copy = 81608 (+55%)

Latency = 55.4 (+74%)

 

Teamgroup Stock DDR5-5200

Read = 64214 (+17%)

Write = 64992 (+23%)

Copy = 58463 (+11%)

Latency = 94.7 (+1%)

 

Teamgroup Tuned DDR5-6200 Gear 1

Read = 91237 (+66%)

Write = 95211 (+80%)

Copy = 85265 (+62%)

Latency = 54.6 (+76%)

 

Teamgroup Tuned DDR5-7600 Gear 2

Read = 99155 (+81%)

Write = 101802 (+93%)

Copy = 91402 (+74%)

Latency = 57.9 (+66%)

 

Conclusions

- Based on the obtained results and input that i saw from other AM5 users online, I am likely held back by the CPU IMC. The RAM sticks did not show much difference between the two kits in Gear 1. Also, I have seen differing results on the X670E Extreme and Hero boards, so that leads me to think that the CPU is my current bottleneck. Curious to see what Zen 5 will bring!

-  As mentioned before, Gear 2 lets the latency suffer a bit, but provides a nice boost to bandwidth. I will stick to my Gear 2 settings for now, as the latency penalty is a mere 3 ns in my case, so nothing to worry about.

 

Overall quite happy with the results at roughly 60-90% performance uplift vs. stock speeds of my previous G.Skill RAM kit 🙂I didnt even really bother testing with the XMP / EXPO profiles tbh, just used them as a quick baseline for additional tuning 😄

 

Aight, with GPU and RAM tuning out of the way, Ill now shift my focus on tweaking my 7950X 🙂 

 

Can you still run DDR4 on these newer AMD 7000 CPU’s? How does really fast/tuned DDR4 compare to faster DDR5 7600? 

[jaybee83]

3 minutes ago, tps3443 said:

 

Can you still run DDR4 on these newer AMD 7000 CPU’s? How does really fast/tuned DDR4 compare to faster DDR5 7600? 

 

nope, as opposed to Intel Alder/Raptor/RPL-R Ryzen 7000 only supports DDR5, no hybrid IMC...

 

My lady's machine sports 12th gen paired with DDR4, curious to see how much RAM OC will improve once I bump her up to 14th gen 🙂 with the 12600 currently installed, she can do up to 3500 in Gear 1 and 3800 in Gear 2.

[tps3443]

29 minutes ago, jaybee83 said:

 

nope, as opposed to Intel Alder/Raptor/RPL-R Ryzen 7000 only supports DDR5, no hybrid IMC...

 

My lady's machine sports 12th gen paired with DDR4, curious to see how much RAM OC will improve once I bump her up to 14th gen 🙂 with the 12600 currently installed, she can do up to 3500 in Gear 1 and 3800 in Gear 2.

Gear (1) low latency DDR4 performs really well. I had a 11900K that could reliably run about 4066-4100Mhz CL14. Back then, people were only running like DDR5 6000 range with 12900K chips. So it was very competitive. 
 

What 14th gen model would you buy to replace a 12600?

[jaybee83]

2 hours ago, tps3443 said:

Gear (1) low latency DDR4 performs really well. I had a 11900K that could reliably run about 4066-4100Mhz CL14. Back then, people were only running like DDR5 6000 range with 12900K chips. So it was very competitive. 
 

What 14th gen model would you buy to replace a 12600?

 

i might as well just go all out with a 14900K 😄 but tbh im not planning to purchase one anytime soon, might wait until 15th gen comes out and prices come down. so far shes quite happy with the 12600.

 

mind you, thats a laptop shes rocking, so naturally the mobo traces and so-dimms are not able to compete with desktop hardware haha

[Papusan]

On 12/5/2023 at 4:12 PM, tps3443 said:

Maybe these specs are true for the RTX4080Ti? If that’s the case maybe it would only be about 20% slower than 4090?
 

 

 

I wonder if there will be a few AD102 dies left for an 4080Ti cards. It seems nvidia have no problem get rid of the defective chips. And the profits is 3-6 times vs put them in an 4080Ti. 

 

Nvidia lists RTX 5880 Ada GPU, presumably a downgraded replacement for the RTX 6000 Ada in China

 

Either way, Nvidia is preparing to release yet another AD102-based ProViz graphics board.

 

Now it's all up to the U.S Government to stop Mr. Jensen and Nvidia from floodind the Chineeze market with loads of AD102 chips. Rather use those chips for the gamers in rest of the world. 

 

On 12/5/2023 at 5:12 PM, Papusan said:

 

4080Ti's if it's real it should have 110 SM's. From techpowerup data page. Thats too powerful if nvidia want to sell new cards to the Chinese market (1/3 of their GPU market). Nvidia was forced to castrate the new 4090D (D for Defective - Downgrade - Disgusting) down to 100 SM's. Why would you create new xx80 GPU's that can't be sold world wide ? And why would you use defective AD102 dies that instead can go directly into the more expensive 4090D/workstation cards for the Chinese market? Hmm... Food for thoughts.

 

 

Today's question.... Will bro @Mr. Fox jump on AMD products within the next couple of years?😁

 

The five worst AMD GPUs of all time: So bad we can't forget them Historically awful AMD GPUs.

 

Dishonorable Mention: Radeon 8500 and Driver Woes

Although Radeon drivers are definitely better today than they were back then, but as with all major chipmakers, AMD (which bought ATI in 2006) still manages to put critical bugs in its drivers from time to time. Just this year, there was a buggy driver that could corrupt your Windows installation, another that transformed Radeon 780M iGPUs into the slower 760M, and even a driver feature that got Counter Strike 2 players banned.

 

 

Quote

More disgusting coming from the Norwegian Government..... 

From the Norwegian Tax Administration

 

There will now be VAT on all online shopping from abroad | The tax authority

 

From 1 January 2024, you must also pay value added tax and any duty on purchases under NOK 350 from foreign online shops. In short.... If you buy an 1 cent screw from an online shop outside Norway and you'll need to pay 25% tax. and of course... Always 25% tax also on top of the shipping.   

https://kommunikasjon.ntb.no/pressemelding/18029125/na-blir-det-merverdiavgift-pa-all-netthandel-fra-utlandet?publisherId=1726411&lang=no

[Mr. Fox]

1 hour ago, Papusan said:

From the Norwegian Tax Administration

 

There will now be VAT on all online shopping from abroad | The tax authority

 

From 1 January 2024, you must also pay value added tax and any duty on purchases under NOK 350 from foreign online shops. In short.... If you buy an 1 cent screw from an online shop outside Norway and you'll need to pay 25% tax.   

https://kommunikasjon.ntb.no/pressemelding/18029125/na-blir-det-merverdiavgift-pa-all-netthandel-fra-utlandet?publisherId=1726411&lang=no

Wow, that's just brutal. It's sad to see examples of a government financially assaulting its citizens in such a malicious manner. It is just so ethically perverse, wrong and messed up.

[electrosoft]

1 hour ago, Papusan said:

 

 

From the Norwegian Tax Administration

 

There will now be VAT on all online shopping from abroad | The tax authority

 

From 1 January 2024, you must also pay value added tax and any duty on purchases under NOK 350 from foreign online shops. In short.... If you buy an 1 cent screw from an online shop outside Norway and you'll need to pay 25% tax. and of course... Always 25% tax also on top of the shipping.   

https://kommunikasjon.ntb.no/pressemelding/18029125/na-blir-det-merverdiavgift-pa-all-netthandel-fra-utlandet?publisherId=1726411&lang=no

 

You now have to worry about anything purchased abroad less than NOK 350 now.....talk about digging in the couch cushions for change....

 

2 hours ago, Papusan said:

 

I wonder if there will be a few AD102 dies left for an 4080Ti cards. It seems nvidia have no problem get rid of the defective chips. And the profits is 3-6 times vs put them in an 4080Ti. 

 

Nvidia lists RTX 5880 Ada GPU, presumably a downgraded replacement for the RTX 6000 Ada in China

 

Either way, Nvidia is preparing to release yet another AD102-based ProViz graphics board.

 

Now it's all up to the U.S Government to stop Mr. Jensen and Nvidia from floodind the Chineeze market with loads of AD102 chips. Rather use those chips for the gamers in rest of the world.

 

This time around Nvidia has zero incentive to lower prices or offer a 4080ti (or any card that uses any version of the AD102)  on the consumer market with the AI market being so lucrative and needing scaled down versions here, there or anywhere.

 

Strix 4090s are now up to $2400 and Gigabyte 4090 at $2000 shipped and sold by newegg....yikes. Basically every card (when they have them as these are the only two in stock) is +$400 over MSRP.

 

 

 

[electrosoft]

8 hours ago, jaybee83 said:

 

nope, as opposed to Intel Alder/Raptor/RPL-R Ryzen 7000 only supports DDR5, no hybrid IMC...

 

My lady's machine sports 12th gen paired with DDR4, curious to see how much RAM OC will improve once I bump her up to 14th gen 🙂 with the 12600 currently installed, she can do up to 3500 in Gear 1 and 3800 in Gear 2.

 

6 hours ago, jaybee83 said:

 

i might as well just go all out with a 14900K 😄 but tbh im not planning to purchase one anytime soon, might wait until 15th gen comes out and prices come down. so far shes quite happy with the 12600.

 

mind you, thats a laptop shes rocking, so naturally the mobo traces and so-dimms are not able to compete with desktop hardware haha

 

The fact it can do G2 3800 bodes well as the memory at least is cleared for take off to 3800 and ~3500 on a 12600 sounds about right but the only way to truly know is to slap it in a desktop board to see. My daughter's 12400 tops out at 3366 DDR4 on her B660 Asus board. My 12900k topped out at 4133 on the Asus Z690 D4 (DDR4) and many 13th and 14th gen chips are sitting around the same level so if you:

 

Get a chip with a known, good 4000+ IMC

Hunt for good silicon / low bleeder (Focus on the 4300 curve)

 

I know both capped to ~130w, this 13900ks runs cooler and scores much higher than my 12900k that's in there now when simulated on a desktop board. When (if?) I switch back to AMD, I should pop this SP115 13900KS in my old  Asus Z690 D4 board and see how far it can go with the B-die sticks that are in there that can do 4500 with 10th gen.

 

 

 

 

[Papusan]

On 12/8/2023 at 12:19 AM, Mr. Fox said:

Wow, that's just brutal. It's sad to see examples of a government financially assaulting its citizens in such a malicious manner. It is just so ethically perverse, wrong and messed up.

 

Yes. The Government didn't like that we could spend up to $32 USD outside Norway and without having to pay tax.

 

On 12/8/2023 at 1:15 AM, electrosoft said:

You now have to worry about anything purchased abroad less than NOK 350 now.....talk about digging in the couch cushions for change....

 

Yup, it's perverse. Even more perverted when you know Norway don't lack money. They have the biggest monetary fund in the world and are one of the wealthiest countries. Our national oil foundation own 2 or 3 % of all noted stocks in the world (5.5 mill people in Norway). We invest money in all other countries (almost everything of $$$), just not in Norway and the greedy state can't even help own citizens....

 

Fattigdom i Norge = Poverty in Norway. This shouldn't be possible with our wealth. But sadly... We see new cases every day... People need to stand in line to ask/beg for free food and milk.

Fattigdom i Norge

 

Food assistance at charity in Norway"

 

Fattigdomsrapport: – Helt hjerteskjærende (Poverty report: - Absolutely heartbreaking).

 

 

Last year, the Food Centers received five million kroner as a "Christmas present" from the government.

 

Despite record-breaking attendance, they have to do without extra support this year.

 

five million Norwegian kroner = Measly 458,463 US dollars to help all the poor living in Norway. Yup, a real joke from the Norwegian government. That was last year. This Christmas.... Nothing. Not a dime for the poor.

 

Yup. This is real in Norway. In defiance of the wealth.

 

On 12/8/2023 at 1:15 AM, electrosoft said:

This time around Nvidia has zero incentive to lower prices or offer a 4080ti (or any card that uses any version of the AD102)  on the consumer market with the AI market being so lucrative and needing scaled down versions here, there or anywhere.

 

There is enough 4090's for sales here home. No shortage of cards but the retail shops follow the market outside Norway. So same price hike here. 

[jaybee83]

8 hours ago, electrosoft said:

 

 

The fact it can do G2 3800 bodes well as the memory at least is cleared for take off to 3800 and ~3500 on a 12600 sounds about right but the only way to truly know is to slap it in a desktop board to see. My daughter's 12400 tops out at 3366 DDR4 on her B660 Asus board. My 12900k topped out at 4133 on the Asus Z690 D4 (DDR4) and many 13th and 14th gen chips are sitting around the same level so if you:

 

Get a chip with a known, good 4000+ IMC

Hunt for good silicon / low bleeder (Focus on the 4300 curve)

 

I know both capped to ~130w, this 13900ks runs cooler and scores much higher than my 12900k that's in there now when simulated on a desktop board. When (if?) I switch back to AMD, I should pop this SP115 13900KS in my old  Asus Z690 D4 board and see how far it can go with the B-die sticks that are in there that can do 4500 with 10th gen.

 

 

 

 

 

some good points right there, thanks bud 🙂 

 

the 3500 / 3800 speeds were with her previous 2x8GB 3800 sticks tho. i have since upgraded her to 2x32GB 3200 sticks, so curious to see how far i can push those. probably not as far, but across the whole lifetime of her machine (im expecting 10yrs) capacity is more important than speed 🙂 plus, of course, its fun to keep upgrading it 😄 

 

im currently preparing aan 8TB QVO 870 ssd for her as a xmas present, shes running out of space on her external 5TB spinner, so thats gonna be a nice bump both in capacity and speed for her. QLC should be fine in her case, were tallking mostly media stuff like movies, music, pics, etc.

[Talon]

New TeamGroup TForce Xtreem 8000 CL38 1.4v 2x24gb kit. Plug and play auto voltages, set to 8400 and set VDD/VDDQ ram to 1.435v. 

 

8400 CL38 easy peazy. 

 

[Mr. Fox]

38 minutes ago, Talon said:

New TeamGroup TForce Xtreem 8000 CL38 1.4v 2x24gb kit. Plug and play auto voltages, set to 8400 and set VDD/VDDQ ram to 1.435v. 

 

8400 CL38 easy peazy. 

 

Congrats. You got a good kit. I RMA'd that exact same kit a couple of weeks ago. It was defective. Couldn't pass any memory tests using the XMP profile with defaults or after giving it more voltage. And, it was unbootable using a saved profile that worked flawlessly with other modules.

 

Edit: I take that back. The trash kit I RMA'd was 2*16GB not 2*24GB. The 32GB kit was rubbish.

Edited December 8, 2023 by Mr. Fox
added correction

[Talon]

16 minutes ago, Mr. Fox said:

Congrats. You got a good kit. I RMA'd that exact same kit a couple of weeks ago. It was defective. Couldn't pass any memory tests using the XMP profile with defaults or after giving it more voltage. And, it was unbootable using a saved profile that worked flawlessly with other modules.

 

Edit: I take that back. The trash kit I RMA'd was 2*16GB not 2*24GB. The 32GB kit was rubbish.

 

Dang, ya I'm done trying my luck with any 2x16gb A-Die kits. The 2x24gb new M-Dies seem to go further. These things can boot 8800 CL38 and I'm trying to stabilize 8600 CL38 right now. Will see if I can get it stable tonight or tomorrow. Otherwise 8400 is worth the price. 

[Mr. Fox]

21 hours ago, Talon said:

 

Dang, ya I'm done trying my luck with any 2x16gb A-Die kits. The 2x24gb new M-Dies seem to go further. These things can boot 8800 CL38 and I'm trying to stabilize 8600 CL38 right now. Will see if I can get it stable tonight or tomorrow. Otherwise 8400 is worth the price. 

Yes, I agree. Now, if (and only if) you can get a good pair of the 16GB modules they are better... much tighter timings and lower latency, but they are hit or miss. I've got the 48GB G.Kill kit dialed in 24/7 stable at 8600. It works great, but my 8200 profile with the 2*16 generic sticks with much tighter timings performs on par with 8600 speed with the looser timings and has lower latency.

 

Seems like everything today is a silicon lottery crap shoot. I am starting to get sick of the amount of nonsense we have to deal with in terms of PC components. There is too much chintzy junk, or grossly overpriced parts; or the deadly combo of chintzy and grossly overpriced junk parts.

 

If I am totally honest, I am also starting to find it hard to identify any major components that I am interested in buying even if I had tons of disposable cash to blow on anything I wanted. This is particularly true of video cards and motherboards. In a very undesirable manner, it is reminding me of the reason I abandoned laptops. There was nothing available that was special or good enough that I would be willing to waste my money buying it. Not quite there yet, but dangerously close. We are surrounded by lots of tacky gamer e-waste designed to titillate the senses of the RGB fangirls.

[Papusan]

On 12/9/2023 at 12:14 AM, Mr. Fox said:

I am starting to get sick of the amount of nonsense we have to deal with in terms of PC components. There is too much chintzy junk, or grossly overpriced parts; or the deadly combo of chintzy and grossly overpriced junk parts.

 

Yup, You're spot on. This video is for our storage junkie @Rage Set I wonder what of the ssd coolers bro @Mr. Fox will go for 😇 Me... Nope. Not in hell. Here is another hot and loud failure. They are everywhere☹️

 

 

 

 

Edit. When we first talk about hot and cold. The cold weather continue here home. Have been over one month with ice cold now. 20 minus and colder. November 2023 was ranked top 4 since 1910 regarding the coldest November month here up in the North. And it won't stop. More on the way... Cold shock on the way: - North wind across the country.

 

 

 

 

[Mr. Fox]

2 hours ago, Papusan said:

 

Yup, spot on. This video is for our storage junkie @Rage Set I wonder what of the ssd coolers bro @Mr. Fox will go for 😇 Me... Nope. Not in hell. Here is another hot and loud failure. 

 

 

 

These products are for a very special market segment. So special, in fact, that Tom wrote a song about them.

---

Interesting metal case fans from Alphacool... I like how they are made but $30 each is too expensive.

 

[Mr. Fox]

A770 has no problem tearing up Gears of War 4 or Gears 5.