I do remember that post now. It’s definitely an outlier and I can’t help but thing there were some confounding issues, perhaps like those I listed above.
[Edit: and that an upgrade lowers the sound quality for a little and if you don’t expect that you may think you are going backwards. I’m talking 5 minutes to 30 minutes, not days.]
For the last N posts I’ve forgotten to mention the other pain in the neck: sometimes the initial boot after an update sounds bad. A power off and on fixes this problem.
There’s a small chance that an update will leave you with a suboptimal set of parameters, in each release we’ve had people sometimes have one channel or both channel volumes get stomped to 0 or some other random value (I’m not talking about the changing to 25 or the bridge bug that changed to 40…) Sometimes the state of the attenuator got stomped, sometimes the limits for how high the volume can go gets stomped… For argument sake there are also things like this that aren’t explicit user modifiable parameters. If any of those kind of errors causes the display/control processor to go into some loop, or read the FPGA a lot more than average, etc. that can mess with the sound quality. This is like the above bad sound on the first boot, but it persists until those internal states are fixed.
There’s also the fact that doing an upgrade lowers the sound quality for a little while afterwards, for me that’s about 5 minutes, some claim longer times.
AFAIK there’s a continuum of different amounts of crud when this happens, but it’s my experience that the chance of it happening to a single person over and over are extremely small so people usually only experience one level of weirdness.
So after listening to Windom for two weeks, I did begin to notice the sound was a bit hot and I would experience some ear fatigue after listening for a while. First I tried dialing down the volume on the DAC to 95, then 90, then 85. At 85 I thought this just isn’t right so I decided to back down to Yale and the go back to Windom.
I have to say that did the trick. The sound is much smoother and any bit of edginess is gone.
Ted says that the flash memory might be getting corrupted. If the thing that gets corrupted is something like brightness, volume, phase then I guess it could get reset. If however it’s some other parameter that the FPGA (or screen) code uses that is not user settable I don’t see how it could ever get reset correctly (unless it’s a bit of code that requires a bit to be zero or 1 and that can get randomly set upon load). I have an active system with three dacs and Snowmass and Windom sound very different from each other. If one of the dacs is always wrongly set after a load that would result in the huge change in sound. I’ve not tried swapping the dacs around, i.e. bass/mid/tweeter - it’s a pain but if Ted thinks it worthwhile I’ll try it.
I had a similar experience with upgrading to Windom. I purchased my DAC about 6 months ago and it came with Snowmass 3.05, so I have never had any earlier versions installed. When Windom first came out, I upgraded from Snowmass 3.06 to Windom, and it appeared that the correct version installed, but the SQ was definitely lacking. I reverted back to Snowmass 3.06, and then upgraded back to Windom with the same results. Based on some of the feedback here, I then installed Redcloud, powered off the DAC twice, then installed Windom, powered off the DAC twice, the version showed to be correct, but it still didn’t sound quite right. Then Ted suggested to someone here to go back to Yale, so I tried the same process going back to Yale, and then to Windom. BIG difference in SQ! Not sure why that worked in my case, could be that going all the way back to Yale cleaned some “gremlins” up in the firmware. However, I truly hope Ted and Paul are able to determine at some point what is happening here. I love my DAC (and all my PS Audio equipment) but the upgrade process really should be more straightforward than this, and we shouldn’t have to hope that we got a perfect load.
First load with stick formatted FAT32 on a Mac - went through a full Load Cycle - blinking light for a longish while, etc. but failed, apparently as it was still Snowmass. Rebooted without drive inserted, still SM.
Reformatted thumb drive to MSDOS-FAT and it loaded fine.
Still kinda “shiny-sounding” to me vs. Snowmass (albeit not have Gone Back Yet), but better off the bat than it is on the Senior-ita ; )
Mr Beef—Really, take an sd card and download/upload Yale. Notice how weak it sounds. Remove card, clean it off, upload/download Windom and see how much different it sounds. It was a happy surprise for my system.
Ron - this is what started the whole Seen-yore/June-yore thing. I am talking about my Jr. No SD cards involved, as it does not have an SD slot. And it is a Different DAC. It is not a 15% thing.
I have both and oddly, the Junior really grabs my attention lately when I listen to it. I loaned out the Matrix that was feeding it and as such I am just doing a USB cable from my desktop computer to the USB input on the Junior. It really sounds very compelling.
At the moment I am enjoying my Senior, and it’s no slouch. The Junior just grabs my attention. Senior is in a regular room setup with my Wilson’s. Junior is on my desk driving B&W 805n speakers.
If it wasn’t such a pain I would bring Junior in here and try to compare the two directly. Instead I will just enjoy which ever one I am listening to.
It would be even neat if the different versions could be in a file format that could be stored on music server and played to do the update of the software making changes fairly painless
There are more than one bug, the Jr is the worst in terms of not getting to the version you want.
If everything were understood, there wouldn’t be any bugs.
I’m not sure that many people repeatedly have updates that get you to the correct version numbers on the screen but don’t sound good. How many people who get something that doesn’t sound good, are carefully checking the version numbers on each update? I’m sure some are, but I’m also sure some aren’t. How many who did an update, got the correct version numbers and bad sound did the extra boot with the rear switch after the update before listening? I’m sure many did, but I’m also sure some didn’t. How many who when having update problems remembered to remove the I2S cables to any non PS Audio devices? How many always used the OS’s eject function EVERY time before they removed the SD card or USB stick after loading files onto it?
Don’t get me wrong, except perhaps for removing SD cards or USB cards without using the OS’s eject function and not removing cables to non PS Audio I2S devices, none of these issues are problems the customer has any control over and none are things customers should have to deal with.
There are enough possible confounding problems that I’m not sure we know the incidence of each of them compared to the others.
Also I don’t think that any of these things has changed since, say, Torreys. Or putting it another way I know that things were worse before Torreys.