Testing RAM with different brands and models, seems to have more difference than I thought

[Windows X]

Testing different RAM with different models and capacity during past few days. They all give different sound signature more than my expectations. Well, I've been doing this for a long while since DDR2 time but maybe my system has become more resolving and better to make more dramatic changes. I'm using bit-perfect playback btw. May consider modifying some like I did for long time ago too.

 

[Macattack]

You don’t perhaps work for a memory chip company? . That’s a new level of audio paranoia I want to avoid.

Please keep reporting!

 

[Windows X]

You don’t perhaps work for a memory chip company? . That’s a new level of audio paranoia I want to avoid.

Please keep reporting!

No I don't lol. It'd be easier to do marketing by asking some memory chip company to build a new batch with own brand, send to some computer audiophiles for reviewing and testing with some raved reviews.

I'm still burning other modules at the moment. At least it'll need some breaking period to get some potentials revealed. So far nothing among what I brought being conclusive yet since they all have different ups and downs.

 

[Windows X]

Here's results I recorded from my mobile phone. I can't record every RAM I tested yet because I have work to do but these two should sound different enough even in mobile phone recording quality.

https://www.mediafire.com/file/698qg6e7j2h4r7r

 

[johndoe21ro]

In your opinion and in your configuration which ones were best?

 

[Windows X]

It has different ups and downs. Corsair is good for wide soundstage and details but focus and dynamics are a bit blur and bloated.

 

[sbo6]

I have heard another big variable in sound is CAS latency. Have you done any experiments as different motherboards mau set RAm timings differently yielding different results per RAM brand and model.

 

[Windows X]

I have heard another big variable in sound is CAS latency. Have you done any experiments as different motherboards mau set RAm timings differently yielding different results per RAM brand and model.

I used to believe in CAS latency affecting sound quality. But if that's really true. All DDR4 will have terrible sound quality because we used to have much lower CAS latency around 3-4 in DDR2. So chip quality and passive parts around it are more important.

I surprisingly found lowest CAS latency DDR4 RAM sounding least preferable with blur dynamics and focus but it has very good airy and expansion though.

 

[sbo6]

CAS in and of itself is one thing, but the continuous increase in memory and bus speed may outweight any CAS latency benefits.

 

[Windows X]

CAS in and of itself is one thing, but the continuous increase in memory and bus speed may outweight any CAS latency benefits.

As over all latency in system increased to support bigger bandwidth, I think optimal timing and latency to get the job done on right time is more important than getting lowest possible RAM.

Regards,
Keetakawee

 

[Felipe Rolim]

Marcin from JCAT and others users (most of them from Audiophile Style) have been wondering about Apacer industrial RAM (https://industrial.apacer.com/en-ww/DRAM/DDR4-Wide-Temp.-UDIMM). Have you been able to try it? I've been using Corsair Vengeance memories for many years (DDR3 and DDR4), but my fondness for them stems from their reliability and durability. I still don't have a good knowledge about sounds. Do you have any indication? What about ECC vs. non-ECC and these Apacer RAM?

 

[sbo6]

As over all latency in system increased to support bigger bandwidth, I think optimal timing and latency to get the job done on right time is more important than getting lowest possible RAM.

Regards,
Keetakawee

Sorry but that's not a true statement. CAS latency and bandwidth are not mutually exclusive. DDR4 CAS latency is higher than DDR3.

 

[sbo6]

Marcin from JCAT and others users (most of them from Audiophile Style) have been wondering about Apacer industrial RAM (https://industrial.apacer.com/en-ww/DRAM/DDR4-Wide-Temp.-UDIMM). Have you been able to try it? I've been using Corsair Vengeance memories for many years (DDR3 and DDR4), but my fondness for them stems from their reliability and durability. I still don't have a good knowledge about sounds. Do you have any indication? What about ECC vs. non-ECC and these Apacer RAM?

I was wondering the same in terms of industrial RAM. I think one of the ultra high end music servers (maybe SGM) uses industrial RAM.

 

[Windows X]

Marcin from JCAT and others users (most of them from Audiophile Style) have been wondering about Apacer industrial RAM (https://industrial.apacer.com/en-ww/DRAM/DDR4-Wide-Temp.-UDIMM). Have you been able to try it? I've been using Corsair Vengeance memories for many years (DDR3 and DDR4), but my fondness for them stems from their reliability and durability. I still don't have a good knowledge about sounds. Do you have any indication? What about ECC vs. non-ECC and these Apacer RAM?

It's OK RAM but not optimal choice for my preference. Technically non-ECC should be more direct but most ECC RAM has better build quality so ECC may sound better in some aspects with better parts.

Sorry but that's not a true statement. CAS latency and bandwidth are not mutually exclusive. DDR4 CAS latency is higher than DDR3.

DDR4 has higher CAS latency which is true. That's why getting lowest possible CAS latency is no longer my priority. In order to provide higher bandwidth, CAS latency is increased. It's been that way if you look back to DDR1/DDR2 where CAS used to be in few single digit.

I was wondering the same in terms of industrial RAM. I think one of the ultra high end music servers (maybe SGM) uses industrial RAM.

If I remember correctly, SGM uses Corsair RAM, 12x4GB of them in Extreme.

 

[sbo6]

DDR4 has higher CAS latency which is true. That's why getting lowest possible CAS latency is no longer my priority. In order to provide higher bandwidth, CAS latency is increased. It's been that way if you look back to DDR1/DDR2 where CAS used to be in few single digit.

I would surmise that CAS latency is still important to sound quality even if it's inversely related to memory technology maturity.

 

[Windows X]

I would surmise that CAS latency is still important to sound quality even if it's inversely related to memory technology maturity.

How do you think about CL3-5 comparing to CL14-19 now? Having lower CL means better sound is no longer true for DDR4. I don't mind if that model will have 1-2 CL higher as long as it's built right and sound right for my server.

I still have my old flagship server with DDR2 running at very low CL. It's all about optimization and people who do better job get better sound. That's why some offer $10-20k servers and they have people buying it.

Regards,
Keetakawee

 

[sbo6]

You're missing my point. Yes CAS latency is increasing as memory tech speeds go up, however, it's still possible and likely that for a given memory type (say DDR4 for example) that lower latency DDR4 memory may improve sound over higher latency DDR4 memory.

 

[Windows X]

You're missing my point. Yes CAS latency is increasing as memory tech speeds go up, however, it's still possible and likely that for a given memory type (say DDR4 for example) that lower latency DDR4 memory may improve sound over higher latency DDR4 memory.

You use the word may and I already explained on that point that it may not sound better. I found a lot of RAM that sounds better even with higher CL for DDR4. Getting lower CL is still more preferable choice if I can choose but sound quality matters more elsewhere for me now. And I wouldn't go blind for lower CL RAM without real testing and comparing seriously first. The best sounding one I prefer doesn't have lowest CL too.

Regards,
Keetakawee

 

[Felipe Rolim]

You talk about "my preference," "my choice," "best sound," but, after all, what memory are you talking about? Which, in your opinion, delivers the best sound and why? Among the Corsair Vengeance, Kingston, Apacer and others, of various series and models, is it possible to identify a sound pattern? Can I say, for example, that a Corsair plays "X" while an Apacer plays "Y"? Or is not this trend?

 

[Windows X]

You talk about "my preference," "my choice," "best sound," but, after all, what memory are you talking about? Which, in your opinion, delivers the best sound and why? Among the Corsair Vengeance, Kingston, Apacer and others, of various series and models, is it possible to identify a sound pattern? Can I say, for example, that a Corsair plays "X" while an Apacer plays "Y"? Or is not this trend?

My point of this thread is different RAM models show significance changes in sound signature and lowest CL RAM doesn't sound better. I already explained in earlier thread that they have different ups and downs. Maybe you missed it.