Will-my-pc-run-Arma3? What cpu/gpu to get? What settings? What system specifications?

[Horus 83]

My 5600X, 4x8GB 3000@3800 16-19-16-36 and RX580 8GB ( U for ultra and S for standard ) 

[Groove_C 267]

2 hours ago, Horus said:

My 5600X, 4x8GB 3000@3800 16-19-16-36 and RX580 8GB ( U for ultra and S for standard ) 

Do you have Aida64?

Your resluts are very low considering your system - 20-25 FPs avg. less in 1080p standard than what I've see other people doing with Ryzen 5000.

My 2014 Intel system with DDR3 is just 10 FPS behind your 2021 system.

 

I think it's because your 5600X doesn't boost much + your RAM is slow, because RAM frequency and just few first timings isn't telling everything.

[Groove_C 267]

With the next 11900K sample I think I will do maxt 5.0 GHz all cores, avoiding 5.2 GHz, even if it's just for 2 best cores out of 8.
Because the last 100-200 MHz require a much higher voltage and thus result in very high temp and power draw.

I hope CPU's memory controller will be able to do 3866 MHz, to compensate for 100-200 MHz lower CPU frequecy, in order to have same FPS.

RAM, when OC'ed, doesn't get nearly as hot as the CPU and doesn't consume nearly as much.

[Mahatma Gandhi 29]

On 8/11/2021 at 9:29 AM, Valken said:

Just comparing GrooveC's 11900K vs 5800x with DDR4000CL16

 

This screenshot for the 102FPS 5800x was very good no doubt. But it wasn't one in a million.

 

I have a 5800x myself meanwhile. 2x8gb Single Rank RAM at 3733MHz CL15, mild CPU overclock (101MHz BCLK), no curved PBO, Beta Bios on a X370 Board, 30 Euro Air Cooler in a mITX System and I reach 100FPS myself.

 

So for sure this Sys was fast, but the statement it was a golden Sample for ArmA3 and a lucky run that is out of range otherwise is not true. What the RKL can do we will have to wait and see. So far I haven't seen one that reached 100FPS+

[Groove_C 267]

7 minutes ago, Mahatma Gandhi said:

This screenshot for the 102FPS 5800x was very good no doubt. But it wasn't one in a million.

 

I have a 5800x myself meanwhile. 2x8gb Single Rank RAM at 3733MHz CL15, mild CPU overclock (101MHz BCLK), no curved PBO, Beta Bios on a X370 Board, 30 Euro Air Cooler in a mITX System and I reach 100FPS myself.

 

So for sure this Sys was fast, but the statement it was a golden Sample for ArmA3 and a lucky run that is out of range otherwise is not true. What the RKL can do we will have to wait and see. So far I haven't seen one that reached 100FPS+

well, only 1 run makes me feel like so + who cares about avg FPS only if it's the min that counts, in Arma, considering how low it can be.

 

And yes, not every 5000 can do 3866 or 4000 MHz and not many of them can boost close to 5.0 GHz.

 

I don't care about 98-101 FPS avg, if min FPS is lower than on Intel, which it is, even if not by much.

[Mahatma Gandhi 29]

But you know my results, so it is proven that 97-100 FPS average is where you end up with a properly tuned poor mans 5800x system with good but not great RAM.

 

And my 5800x I got cheap, because it was the worst in a binning done by a few folks I know. But you know all that. We talked about it.

[Groove_C 267]

10 minutes ago, Mahatma Gandhi said:

But you know my results, so it is proven that 97-100 FPS average is where you end up with a properly tuned poor mans 5800x system with good but not great RAM.

 

And my 5800x I got cheap, because it was the worst in a binning done by a few folks I know. But you know all that. We talked about it.

Again - avg is not relevant for A3. You can have it as high as you want, but if min FPS is low/lower, who cares if avg is 95-97 or 98-101.

 

You won't see/feel a difference between 90 or 100 FPS, but you will when min FPS is low/lower.

 

Also the lower the settings (resolution/quality), the "better" Ryzen is, due to it's cache. But nobody plays at such settings. Once you go for 24/7 settings, the picture is different.

 

And since you also have a 1070 Ti, like me, we will benchmark at Ultra settings, once I get the new 11900K.

 

Lets maybe compare max FPS and 720p and very low settings, where Ryzen will be even better, but again, nothing to do with real 24/7 settings.

[Mahatma Gandhi 29]

Sure, I fully agree.

 

My point was, the 102FPS documented have not been a particularly obszene result for a 5800x as I get 97-100 with a bad sample, SR RAM, an ITX box, and a 30 Euro cooler. That was the only thing I wanted to add to the discussion.

 

Min FPS for me are about ~71/72 on average.

 

And by the way, I am hard limiting mine to 95W power draw.

 

EDIT: Also the lower the settings (resolution/quality), the "better" Ryzen is, due to it's cache. But nobody plays at such settings. Once you go for 24/7 settings, the picture is different.

 

Thats the first time I hear that. Do you think so or do you have data on that?

 

 

[Groove_C 267]

6 minutes ago, Mahatma Gandhi said:

Sure, I fully agree.

 

My point was, the 102FPS documented have not been a particularly obszene result for a 5800x as I get 97-100 with a bad sample, SR RAM, an ITX box, and a 30 Euro cooler. That was the only thing I wanted to add to the discussion.

 

Min FPS for me are about ~71/72 on average.

 

And by the way, I am hard limiting mine to 95W power draw.

 

 

Well, power draw is critical for some and not so much for others.

11900K was doing few min FPS more and that's without tuning RAM to the max possible, since I had the CPU for only like a day or less and only 3733 MHz, not 3866 MHz and no 100 MHz BCLK increase to something like 102 MHz.

[Mahatma Gandhi 29]

But money matters for most people 😉

 

 

[Groove_C 267]

4 minutes ago, Mahatma Gandhi said:

But money matters for most people 😉

 

 

when it's about money, 5600X is the only possible candidate.

 

Maybe 11600/11700K as well, since they cost less than 5600/5800X, but I think they will have problems with IMC and will require a very high voltage (temp/power draw) for target frequency vs. 11900K.

[Groove_C 267]

@Mahatma Gandhi but then again, 5600X is the lowest quality there is, among X CPUs.

So the chance that it will also need a much higher voltage for a given frequency and the max frequency won't be that high + Infinity Fabric quality...

[Mahatma Gandhi 29]

There is no logical explanation for such a difference. They disable two cores because one might have an error, but that does not necessarily mean that other parts of the chip are affected. Same with Intel. I don't see a reason why lower tier models should not reach the same memclock or uncore freq. In fact, I had a 3300x CPU that had no problems running FCLK 1900+ and 3800MHz RAM.

 

And in most cases, they disable cores not because they are broken but just to diversify the products. There are more 5600x sold than 5800x, yet there are for sure more fully functional dies coming from the fab then ones with defective cores.

 

Last not least, the fastest on the current list is this one: https://imgbb.com/0tSC7XB

 

And I want to pick up your statement again that Ryzens are slower at high quality settings. Do you make that up or do you have proof? Its the first time I hear that.

 

 

 

 

[Groove_C 267]

22 minutes ago, Mahatma Gandhi said:

There is no logical explanation for such a difference. They disable two cores because one might have an error, but that does not necessarily mean that other parts of the chip are affected. Same with Intel. I don't see a reason why lower tier models should not reach the same memclock or uncore freq. In fact, I had a 3300x CPU that had no problems running FCLK 1900+ and 3800MHz RAM.

 

And in most cases, they disable cores not because they are broken but just to diversify the products. There are more 5600x sold than 5800x, yet there are for sure more fully functional dies coming from the fab then ones with defective cores.

 

Last not least, the fastest on the current list is this one: https://imgbb.com/0tSC7XB

 

 

 

 

 

 

There are 2 types of CPUs. Those, that don't match the requirements to become a certain model and those, that match all the requirements, but the market says it's too expensive and too many cores and the customer doesn't have so much money and doesn't need such a CPU, so let's disable some stuff, but quality of what remains, remains the same, top quality.

But for different models, there are and always have been different voltage for frequency and IMC requirements.

 

It's like when you buy a 5950X, the probability that you won't be able to undervolt it as much as other, lower models is much greater, because it already requires less voltage for same or higher frequency, so you can't downvolt as much, because the default voltage is already low enough.

And this is what you are paying for. You're not paying only for more cores/cache, but also for better quality chips, on average, not all of them.

[Mahatma Gandhi 29]

I have certainly not been living under a rock and as said and as you agree, there are CPUs that are artificially reduced in core count which are just the same as the top tier models and then there are the ones where a core is defect and where those are used to produce lower tier SKUs.

 

But there is certainly none which are selected based on the IMC or uncore frequency capabilities. The max spec is 3200MHz and all of them work at that. OC is OC and they dont bin for RAM OC. Neither of them, AMD or Intel.

 

Core Voiltage is also identical between 5600x and 5800x. Two CCD Models might be selected at that end, but certainly not the lower ones. 

[Groove_C 267]

4 minutes ago, Mahatma Gandhi said:

I have certainly not been living under a rock and as said and as you agree, there are CPUs that are artificially reduced in core count which are just the same as the top tier models and then there are the ones where a core is defect and where those are used to produce lower tier SKUs.

 

But there is certainly none which are selected based on the IMC or uncore frequency capabilities. The max spec is 3200MHz and all of them work at that. OC is OC and they dont bin for RAM OC. Neither of them, AMD or Intel.

 

Core Voiltage is also identical between 5600x and 5800x. Two CCD Models might be selected at that end, but certainly not the lower ones. 

Well, 11900K is also binned, like 5950X, unlike the lower end models.
That's why I bought it, for lower stock voltage, to be able to have lower temp/power draw at same or even higher frequency, since I'm on air + I want to extract everything possible from the architecture, considering the cooling I have.

[Groove_C 267]

14 minutes ago, Mahatma Gandhi said:

But there is certainly none which are selected based on the IMC or uncore frequency capabilities. The max spec is 3200MHz and all of them work at that. OC is OC and they dont bin for RAM OC. Neither of them, AMD or Intel.

 

Core Voiltage is also identical between 5600x and 5800x. Two CCD Models might be selected at that end, but certainly not the lower ones. 

Well, Intel has always had differences in IMC max capabilities between the models and one can find a lot of proof of it.

 

i7 CPUs, when i9 didn't even exist, could hit higher RAM frequencies than i5 CPUs, on average and that with the same RAM kit and mobo and that for dozens/hundreds of CPUs manually binned.

[Groove_C 267]

And I never said that 5800X is supperior to 5600X. I said only that 5600X is the lowest quality AMD has and it's true, vs. the top models.

[Groove_C 267]

Sure, there is no binning for 11700K, 11600K and all the others, but that's also the reason why I don't buy them.

[Groove_C 267]

4 hours ago, Mahatma Gandhi said:

EDIT: Also the lower the settings (resolution/quality), the "better" Ryzen is, due to it's cache. But nobody plays at such settings. Once you go for 24/7 settings, the picture is different.

 

Thats the first time I hear that. Do you think so or do you have data on that?

Like I said, the picture is different once you use 24/7 real settings.

I've already seen games' tests, where depending on resolution and video options, Ryzen 5000 traded places with Intel.

[Groove_C 267]

So since we both have a GTX 1070 Ti, lets test Arma at real video settings, like very high and not just at standard settings, once I have my other 11900K, even if I always play at 1440p ultra.

[Groove_C 267]

Of course, only slightly better min FPS with 11900K vs. 5800X comes with a price and power draw that many will find very hard to justify, no doubts, but everyone buys different stuff.

 

Some want min FPS to be as high as possible, for what they find ok to spend money on.

Others simply want the highest min FPS possible and can spend substantially more money to achieve this and it's also ok.

[Horus 83]

For 4x8GB this is a fine memory and cheap

 

[Groove_C 267]

1 hour ago, Horus said:

For 4x8GB this is a fine memory and cheap

 

This is exactly what I was thinking about and wrote about it.

Almost 60 ns in Aida64 and that's after optimising + relatively low 5600X frequency.

That explains your 20-25 lower FPS vs. some other people also with Ryzen 5000.

[Horus 83]

On 8/18/2021 at 9:14 PM, Groove_C said:

That explains your 20-25 lower FPS vs. some other people also with Ryzen 5800x.

Small correction                                                                                                               ☝️