Our new system. Have I made any mistakes?

[Scott]

All the high scores on the leaderboard are all OC machines to the max being cooled by liquid nitrogen, PC stood in a freezer with 40 fans aimed at it.
The clock speeds on the gpus are a lot higher than the factory oc gpu I have, memory frequency too. Plus whatever else they are pushing to the max.
Mine is out of the box how PCS set it up, nothing changed whatsoever.

I'm not the most clued up on PCs in the regards to the CPU speed you mention, all I can say is it generally hits those speeds when benchmarks are run, and has gone as high as 5011 mhz
The CPU runs hotter than the GPU, think its around 60-65 during timespy stuff.
Long gaming sessions say 5-6 hours and at 4k highest settings possible that are suitable the Cpu can go into the high 70s but it tends to be briefly, like a spike kind of thing like when a game loads up,probably averages in the mid 60's.

The Gpu rarely goes into the 50's.

This is with the fans running pretty silent.

The really top scores are, I was referring to the scores with the 5900X and the 3080 pairing though. Where it appears to me that all the setups are fairly standard. For extreme setups people won't often choose an AMD chip.

Here is an example running the 5900X without an OC and it looks like the 3080 Strix without an OC as well. The frequency values are all fairly typical around the 18500 score mark. Overclocked rigs tend to score 19000+.

5Ghz from that chip really is fantastic.

I scored 18 500 in Time Spy

AMD Ryzen 9 5900X, NVIDIA GeForce RTX 3080 x 1, 32768 MB, 64-bit Windows 10}

www.3dmark.com

You can have a play around here if you're interested...

Search

www.3dmark.com

You set the CPU to what you are running, the GPU to what you are running and be sure to select the count of GPUs as 1.

 

[scarpa1]

The really top scores are, I was referring to the scores with the 5900X and the 3080 pairing though. Where it appears to me that all the setups are fairly standard. For extreme setups people won't often choose an AMD chip.

Here is an example running the 5900X without an OC and it looks like the 3080 Strix without an OC as well. The frequency values are all fairly typical around the 18500 score mark. Overclocked rigs tend to score 19000+.

5Ghz from that chip really is fantastic.

I scored 18 500 in Time Spy

AMD Ryzen 9 5900X, NVIDIA GeForce RTX 3080 x 1, 32768 MB, 64-bit Windows 10}

www.3dmark.com

You can have a play around here if you're interested...

Search

www.3dmark.com

You set the CPU to what you are running, the GPU to what you are running and be sure to select the count of GPUs as 1.

I had looked their already.
Even those scores have OC speeds and memory frequency.
Might not be like the extreme ones but still a good bit higher than out of the box.
I'm pretty sure the memory frequency makes a bit difference looking at the scores, even small amounts add up to big gains.
I gained loads of points just from my ram settings as when I first ran a Timespy test it was pointed out it was only running at 2133mhz , think it may have been yourself, so even slight OCs on the gpu will gain plenty.

 

[VikCha]

I would strongly recommend the following:

Hero Motherboard
500GB 980 Pro M2 drive
1TB 980 Pro M2 drive
1000W RMx PSU

The OC will likely work in your favour given your uses so should suit well, be sure to benchmark/check your throughput with the OC on and off though, just to be sure.

This is one of the few cases where a build like this would actually be recommended. When spending the cash like this you might as well get it looking sweet too

Thanks for your time.

Can I ask what difference the Hero upgrade will make.

Is the power supply not enough to handle this GPU? or do you recommend having some capacity left over for future upgrades?

Many thanks to everybody for your advice and input:

Vikesh

 

[Ash_]

Hero can handle heat better and more features, therefore, it won’t throttle, even under heavy loads.

Psu, these things can last in excess of a decade, ask @SpyderTracks how old his is. You need to think into the future and what you might have in X many years. Also, when you drive a car, you don’t keep it at 5000rpm, same applies with a psu - keep it nicely loaded and it will last longer. Well worth the investment in a psu, especially with high powered kit.

 

[SpyderTracks]

Hero can handle heat better and more features, therefore, it won’t throttle, even under heavy loads.

Psu, these things can last in excess of a decade, ask @SpyderTracks how old his is. You need to think into the future and what you might have in X many years. Also, when you drive a car, you don’t keep it at 5000rpm, same applies with a psu - keep it nicely loaded and it will last longer. Well worth the investment in a psu, especially with high powered kit.

Mine is a Corsair HX750w and I've had it in 3 separate builds since 2009, it's incredible.

It is really worth getting a decent PSU, especially if you've got higher end components in there, it's dictating the quality of the current and voltage you're sending through every single component in your rig. Very important critical component in the build and often overlooked.

 

[Ash_]

The really top scores are, I was referring to the scores with the 5900X and the 3080 pairing though. Where it appears to me that all the setups are fairly standard. For extreme setups people won't often choose an AMD chip.

Here is an example running the 5900X without an OC and it looks like the 3080 Strix without an OC as well. The frequency values are all fairly typical around the 18500 score mark. Overclocked rigs tend to score 19000+.

5Ghz from that chip really is fantastic.

I scored 18 500 in Time Spy

AMD Ryzen 9 5900X, NVIDIA GeForce RTX 3080 x 1, 32768 MB, 64-bit Windows 10}

www.3dmark.com

You can have a play around here if you're interested...

Search

www.3dmark.com

You set the CPU to what you are running, the GPU to what you are running and be sure to select the count of GPUs as 1.

Just to add to this, i just benched my PC (couldn’t do it thursday as i work away) and my 5800X maintains 4.85ghz the entire bench and sits at roughly 46 degrees on a H150i, although my GPU did hit up to 70 degrees.

With enough cooling, the Ryzen 5000 series will maintain outrageous boost clocks

 

[SpyderTracks]

Just to add to this, i just benched my PC (couldn’t do it thursday as i work away) and my 5800X maintains 4.85ghz the entire bench and sits at roughly 46 degrees on a H150i, although my GPU did hit up to 70 degrees.

With enough cooling, the Ryzen 5000 series will maintain outrageous boost clocks

That’s the thing with Ryzen 5000, they actually UNDERSTATED the capabilities and the advertised boosts are with a very poor chip, in general, they exceed advertised clocks. But you have to have decent cooling for them to reach optimal clock speeds.

 

[Ash_]

That’s the thing with Ryzen 5000, they actually UNDERSTATED the capabilities and the advertised boosts are with a very poor chip, in general, they exceed advertised clocks. But you have to have decent cooling for them to reach optimal clock speeds.

With MSI’s ‘Gameboost Mode’ on... the 5800X stock speed is 4.3ghz. I’m going to bench with that now and see the impact. I know we both agree though, overkill cooling is not even really overkill once you’re over 6 cores and is entirely worth it

 

[SpyderTracks]

With MSI’s ‘Gameboost Mode’ on... the 5800X stock speed is 4.3ghz. I’m going to bench with that now and see the impact. I know we both agree though, overkill cooling is not even really overkill once you’re over 6 cores and is entirely worth it

I’ve always been a big believer in overkill for both the CPU /GPU coolers and the case / fans. I spend a ludicrous amount of money on cooling, but firmly believe it’s well worth it.

 

[Ash_]

Okay gameboost mode literally held 4.3ghz the entire bench. Not sure what the point in it is, maybe it acts differently in game versus benchmarking.

With the correct cooling though, these chips are extremely capable and a decent cooler is essentially an upgrade to the chip