I made sure my two cards, (both 32gb V30 one SP the other Kingston) were on v5.51 xx.212 and set to 144mhz (they weren’t matched previously)
I swapped them back and forth several times (what a faff ) each getting 5 listens to the same reference tracks
the SP card was slightly better!
The difference is soundstage; it’s more constrained with the Kingston and it opens up with the SP - it’s not night and day, but definitely preferable with the SP
I’ll double check later, but if the findings remain the same I’ll be happy to try the more expensive Delkin card…
Do you consider the differences obvious? I.e. would you be able to spot it blindly?
And, what would you say could possibly make this difference? @Marin_Weigel is fully correct, the system only loads into RAM from the SD-storage, and i dont even think system logs to storage, only to virtual drives in RAM.
The only thing i could attribute to any change could be power consumption, if that varies between different cards…
And, better buckle up, the ASR crowd is gathering and getting ready to attack.
Bearing in mind I tried before (with other cards and these two with different settings) I was surprised to hear anything
Of course we’re all aware of bias, hence I swapped them Several times with different tracks played - and I slept on it - to return with fresh ears the next day
You could easily use some virtual loopback device on either Windows, Linux, or Mac to record both versions of the digital stream and then do a null test with free software to see if there really is a physical difference … but I‘d bet that‘ll never happen, right?
Just tried the cpu frequency adjust set to 528 MHz and I did not like the change in sound, in my system, to my ears, it ‘thins’ out the midrange emphasizing deep base and top end. Sounds becomes a bit more bright and this brightness adds to the impression of a more detailed bass. But the cost is midrange fullness and naturalness.
In my system, the cpu latency options has the same effect
Interesting, I had very similar results, especially in the lowest bass region - so I kept pushing the CPU frequency down to 144. I found the changes very beneficial
I had to tame the bass, but couldn’t do it with RoomPerfect on my amp as its source related. I was expecting to have to move seat/speakers and/or change the voicing and/or crossover in the amps (Lyngdorf ‘3400 and ‘2200)
Coincidentally, I’d made some abfusers. Placing two on the side walls pretty much resolved the bass, and subsequently one on the ceiling improved things further
On the up side, at least you can put the CPU back to Performance to return your preferred sound
It looks like Sotm have released another upgrade with additional options to tweak- also they appear to have fixed the ‘locking up’ issue when running lower cpu speeds - sounds great!
I seem to get the impression these CPU and Latency tweaks have a significant impact on the sound quality. Unfortunately they also seem to affect Roon playback in a negative way, causing pitch variations and glitches on the SOtM when using grouped Zones AND upsampling to the SOtM based zone, in Roon.
I am running these settings and it sounds really good.
Except for when i group this Zone with another Roon Ready Zone, when audio freaks out completely. It seems there is a network overhead with grouped zones, to better sync between them?
Without looking (I don’t leave my system on, apart from the clock) I’d say my settings are the same, or very similar. And yes, they do impact the SQ and the ‘right’ one is beneficial on my system too
I don’t group zones but I am using the upsampling. The only thing I’ve noticed is it doesn’t like DSD; it can lock up after a while and needs a reboot of the SMS. Not that this is a major problem for me as I only have a few tracks ‘donated’ to me for comparison purposes. Otherwise it sounds great
I guess with mine, the low SOtM CPU speed is somehow overwhelmed by DSD, causing a lock up? What I don’t understand in your case, is whether the Roon server is struggling and mi-feeding the SOtM, it then playing what it’s receiving, or is Roon providing a ‘clean’ signal which the SOtM is somehow corrupting?
I’ve no proper understanding/explanation of either of our issues, but intuitively it would seem the Roon server (or, as you say, network) is struggling in your case?
Ah, no, not the Roon Core at all.
I upsample to max PCM (Even multiplier) and run the Denafrips Ares II in NOS mode.
The Roon Core does this with the selected filters at a processing speed of 60x, regardless of the material, and only slightly display CPU activity in Activity Manager (Win11).
And any anomalies while grouping are alleviated by setting the SOtM to 1000Base-T instead… But it doesn’t sound quite as good that way! So i prefer to not group this Zone, and keep the 100Base-T mode!
) each getting 5 listens to the same reference tracks
