Of course they could just stop offering free updates to all of us and apply updates to new sales only.
Lol. Nice. But agreed. Here is what I appreciate. Honesty. Yep we have a bug. And yes it is not yours ted. Doing development work - and hearing business say ‘we have a bug!!’ Used to bug the hell out of me. Pun intended. Thanks for stepping up ted.
1 Like
Ah, the soup nazi approach 
Paul is an Apple fan and appears committed to their model of selling hardware with software updates free. It’s a good model.
Of course.
Lemme ask though. Any possibility this is a hardware issue? I’ve never had an issue (no jinx). Might serial numbers help at all?
At a company I worked for about 35 years ago we had “random” crashes in the field but no idea what was going on. At a certain point we had an edict that if one of those crashes happened everything should stop on that machine until I could gather all the info I could. After about a week I realized that in every crash there was a hex 80 somewhere where it wasn’t supposed to be (in a register, on the stack, in ram recently read, etc.). It became the “80’s from space” bug. Eventually we got to the point that we could reproduce it about once per day on a particular machine in the lab. I lived there for the next month. I finally noticed that every time an 80 from space happened that all pages in memory had an 80 at the same location. After staring at the wall thinking of every place we ever could write an 80, I figured out that the test and set instruction was probably doing it. The hardware guys said “don’t do that.” So I didn’t and the bug was fixed.
1 Like
Nothing points that way. The DS and the DS Jr share no hardware in that part of the code. But they share the update code, etc. If it comes down to that they certainly have a bunch of people here that they could ask for serial numbers.
Ok cool. In my experience - its not the first bug that gets you. It’s the second one you introduce by making a ‘fix’ while making a poor assumption. Then everything really goes downhill fast. People don’t understand that. One of the reasons I oinked at Paul about introducing features (auto decrease in volume on startup because someone didn’t check volume on delivery) while the team was trying to fix bugs. You don’t do that. 
Odds of introducing more bugs increased dramatically. Patience Yago. Maybe you get that
Good 'ol 360
[Edit - For those of you not familiar with “The Mythical Man Month”: At a certain point IBM recognized that they introduced 1.x bugs for each bug fixed in OS360… It was time for the 370.]
I ask this genuinely… How many dac’s this old near this price point sounds as good as the DS DAC sounds?
It took an extra 10 mins of hassle to fix the bad load. I was legitimately thinking about getting a new DAC a week or two ago. Now I’m not. I’ll admit that this bug still lingering around is annoying… But I think it’s genuinely small potatoes.
5 Likes
That really isn’t the point though, is it? It’s not like the DACs were released a few months ago and this a new problem. I have only been around for three updates and each one is creates a bunch of problems. If it were my DACs, I’d be fixing the problem instead of living with it. It’s not like it is a minor problem that no one sees. But hey, that’s just me…and I have never experienced the problem myself.
For me, its not as simple as a certain percentage of users have a problem, and we address those individual cases. Its that no can really know if they have a problem or not. There’s no current way to empirically verify that a load is ok. For me, multiple loads have created a variety of different outcomes, and I have no idea which one is how Ted intended. I keep pulling the slot machine lever. While certainly minor in the scheme of life, it does create doubt about my beloved DAC.
In that light, this issue effects every single owner of a DirectStream DAC. Those that have a problem, those that think they have a problem and don’t, and those who are concerned they might have a problem so they run through every update scheme and in doing so, may have created a problem that didn’t originally exist. It effects those that are oblivious to all of this who may have a problem but not realize it. Keep doing the math, it stretches quite far and is about the integrity of the user experience.
I say all that having zero idea what’s involved with creating a real fix, or even if one is possible.
1 Like
You certainly have a point. I don’t intend to imply you don’t.
I just don’t see this as that big of a problem overall. I concede completely that for this kind of money there are some that would be put off by this. I also concede that for this kind of money I don’t know if any other dac that has had this long of a life cycle that is, despite these bugs, much better than it was at first release 5 or more years ago.
Shoot, I heard Redcloud for one song and couldn’t handle it. Might have been a bad load, but if that was the dac’s peak performance I wouldn’t even own it. If ps audio continues to spend money on firmware upgrades and never fixes this bug… I’ll still be more than happy. And maybe I’ll hold onto it long enough to end up with a TSS dac.
1 Like
I answered the first issue: upgrade from two different earlier releases (making sure the version is correct on all four updates.) If the sound is the same you are done. If it isn’t do it again from a different old release. Most can simply notice if the update to Windom is better than Snowmass and if not upgrade from an earlier release.
Not that we want anyone to have to do any of that, but I highly suspect that people may not be checking the version numbers after each change of OS. If not they may not know if their update happened at all (which is a separate problem.)
This is far from ideal, but it works for almost everyone. FWIW There’s nothing wrong with preferring Snowmass (or any previous release.)
I.e. The problem, tho not great, is far from as bad as you see it.
3 Likes
I’ve followed all of this pretty closely, and this sounds like a new recipe! By “upgrade from two different earlier releases” do you mean go from the starting point of like Snowmass, which I assume is update #1 in your 4 update formula:
- Load Huron. Check to makes sure its there.
- Remove USB drive and reboot.
- Load Yale. Make sure its there.
- Remove USB drive and reboot.
- Load Windom. Make sure its there.
- Enjoy Windom.
Is that right? Hadn’t heard the “update from two previous releases” part.
The overall PSA and DirectStream experience is wonderful. This one part of the experience is wonderful in the same way I imagine owning 30 cats would be wonderful 
For all we know, conversations in the Oval Office could’ve gone like this: “Sir, Turkey wants to attack the Kurds, what do you want to do?”. Trump: “Whatever, I’m busy with my stereo. Hey, take a listen, does this sound like a good Windom load to you?”
The future of the free world could rest on your shoulders Ted!!!
Finally got Windom installed this evening on my DSJr. I guess my unzip utility does something weird to the files. James from PS Audio emailed me the files and it worked fine.
Now I’m listening to Anour Brahem’s album “The Astounding Eyes of Rita” which is one of my go to albums for new gear, and just to relax in the evening. All I can say is “wow.” This update is like taking a component to another level. More tone, texture, detail and spatial depth and width. I can see where some may feel it’s too “lush” if they already have a softer sounding system, but not me. I think this is perfect. The rest of my system is a PrimaLuna Dialogue One integrated with Sophia Electric EL34’s and Salk Song3 speakers streaming Tidal via Roon.
2 Likes
Not quite: you can do whatever loads you want, but you may need to come from x to Windom, and if that’s not what you want come from y to Windom. I’d recommend Yale and Huron based on the posts I remember with various releases.
1 Like
In my company, we did once experienced random FMS (Flight Management System) crashes.
A senior engineer spent ~ 1 month (on and off) analyzing the FMS memory with a bus analyzer to see what was going on.
He eventually found out that sometimes the address output of the RAM chips would flip one bit for no reason.
He started looking at possible service bulletins for the RAM chips we were using and sure enough he found one where the manufacturer was warning about that very problem with a list of part numbers for the affected chips.
However, the purchasing employees had failed to read the service bulletin and we ended up producing a significant number of FMS units with faulty RAM chips, which resulted in a bunch of recalls from our customers.
Not great.
So yes, H/W problems can happen, but hopefully not in the PS Audio DACs.
1 Like
Have you personally had this issue with updates to your psaudio DAC?
Please read that post you are responding to more closely.