My Bridge 2 died with the previous upgrade:(( Still non functional!
Snowmass loaded like a breeze and sounds great, even a little bit to pronounced bass, maybe it’s my room. No wake up/sleep problem. Remote is working as it is supposed to be.
In the meantime I also experienced, bass can be massive in certain recordings.
If this is flat, Redcloud wasn’t.
I installed Snomass 3 days ago.
I didn’t think the DS could get any better after installing Redcloud.
Snomass represents the most significant improvement yet.
Every facet of of the sound is better.
Period.
1 Like
It’s less likely to be an amplitude difference between Redcloud and Snowmass. Snowmass has better time alignment (more accurate phase response.) The operative differences between Redcloud and Snowmass are the PCM upsampling filters and all such filters I use are quite accurate down to DC. Lowering FPGA noise can lead to changes in perception, but things like average amplitude aren’t affected.
I read in one of Paul’s posts that tedsmith will be working on a “cost no object” dac.
What kind of things could you do to actually improve on the DS?
And what is a TSS?
I guess this is why everything sounds better as soon as one compensates the new situation with adjusting the bass level compared to the previous situation. As it also sounds rather like my vinyl setup, it was a win win.
TED Smith Signature DAC. The nickname for a 20k range new development in progress (see another thread)
As jazznut said:
https://forum.psaudio.com/t/tss-two-chassis-super-dac/6974
A few posts in are links to other TSS threads.
It’s interesting to me that my first impression of Snowmass I posted after half an hour listening was much more to the point imo than it would have been after longer listening.
Anyway, isn’t it fascinating how much better emotional experience you can give to a lot of people who care about music with your „number crunching“ 
1 Like
Despite how much grief I give people that show up and claim wire is wire, bits are bits, etc. I do use technical specs, simulation, numerical analysis, FFTs, test tones etc., and when they say something is better almost all of you agree that it is better.
I didn’t necessarily expect that to be true when I started. Now I can pretty reliably design something and know that it will sound better without listening (tho, of course, I do.)
3 Likes
Output transformer looks great. How does the inductance compare with the DS?
If you literally mean inductance of the transformer proper: I use the leakage inductance as a part of the output filter. The mutual incuctance defines the bandwidth: with a 600 ohm drive and a 600 ohm load the -3dB bandwidth of the transformer is 0.4Hz to 200kHz. The DS transformers aren’t speced as precisely, but I suspect that they are flat from, say, 4 or 8 Hz to 100k or 150kHz.
If you meant the output drive impedance I’ll still be around 120 ohms flat across the audio band.
Thanks. Yes, I was just curious of overall inductance to get a general idea of how it would compare with the DS in terms of bass response. The pdf specs suggest it should sound impressive.
That’s interesting:
I get that it’s always better sounding in many characteristics then, but when e.g. Redcloud was better than the predecessor but still had no optimal phase response, which when better corrected in Snowmass lead to very different tonality and more realistic performance…isn‘t it that in this case a new version may sound better than the old one in many specific characteristics but may sound more „wrong“ than a previous ones tonalitywise?
Just like many now seem to have the impression, Redcloud sounded better than everything before but also many say Redcloud didn’t have the tonality they prefer now in Snowmass and certain pre-Redcloud versions?
Isn’t it kind of luck, how tonality turns out when technical parameters improve but are not perfect yet?
Not necessarily luck. If you look at my predictions in August about what the new release would sound like I pretty much called it before I’d written a line of code. I’ll grant you that what our ears make out of certain FPGA noises seems to be pretty random, but from release to release I’m lowering the FPGA generated noise and that’s not random.
I see that you could predict that e.g. Redcloud had its tonality and Snowmass another tonality by the respective improvements. But do you know which one’s closer to the final goal and „truth“ of a „perfect“ release?
For me that’s just hypothetically as in case I can adjust my speakers to each releases tonality and just listen for the other characteristics, so I always just have positive experiences fortunately.
(Except for Yale (or was it Pike’s Peak?) and except for bugs) Each release has uniformly been technically better than the previous one. Snowmass is more faithful to the input than was Redcloud, which was more faithful than Huron, … Each of these releases had a major measurable technical step from the previous release.
I.e. I’m not wandering around in the wilderness. Feedback from Paul, Darren, (Arnie in the past) and many of you help me learn (roughly) what technical differences cause what sound quality differences. But at times it takes a little while to figure out how to implement them within the confines of the current hardware.
3 Likes
Thanks Ted, I didn’t want to doubt anything you mentioned in this post.
So for DirectStream 2 (not TSS), which I assume you’ve at least thought about, what hardware could be changed or upgraded while keeping the device at the same price point? A newer model FPGA obviously, but is there any other low hanging hardware fruit ripe for upgrading or tweaking?
Ted,
Are you saying that the TSS will have even lower output voltage level than the DS, or is it the noisefloor you are referring to?