Paul McGowan said
It is essentially the same, cleaned up and all put together. I gave the beta testers a simpler version but the same from a sonic standpoint.
This is embarrassing to admit, but I do not know how to download, unzip and then transfer a file to an sd card. Will this update be available to purchase already loaded on an sd card? Thanks.
We can walk you through it… what kind of computer do you have?
Windows 8 desk top pc
Just follow the link that Paul posted in his previous post. Then download the file. One of the files will be a PDF that explains what to do… but simply as long as you have a 2gb SD card which you should have, just pop in it your computer’s SD card reader, remove the existing files put the new ones on and then install the SD card into the DS as the PDF instructs… If you really can’t do this then I’d call PSA and let customer service walk you through it… it’s not difficult.
As you form opinions, please post your impressions. We had a productive discussion in beta and I am curious to learn what others hear (without prompting with my impressions).
I have had my DS upgrade for a week. 100 hour of running. Very impressed with it straight out of the box and it has continued to improve noticeably over time. Loaded the new firmware today. The new firmware is another step altogether. Female vocals are incredibly real. The air, life and musicality of the new sound is exquisite. Very very pleased!
Thanks guys! Glad you all liked it. Great when we can make changes with a product and give them to our owners for free. Just like Christmas only it happens in the summer!
Paul mentions “letting it settle” when upgrading to the new firmware…? Is there an expected break-in for firmware?
not sure of the firmware naming convention - the new one is 6115 apparently, which seems to be a mash of CPLD 0.61 and version 1.1.5. So the original is CPLD 0.54 version 1.1.2 should be 5412.
I listened to 5412 for 30 minutes, then 6115 for an hour, then back to 5412 for 30 minutes. I prefer 5412. It has less edginess, it is more subtle, fluid and is more musical. 6115 sounds more like the PWD2 to me, with a harder, more pronounced leading edge on notes.
Perhaps I need to let 6115 settle for a longer period. If so, someone please explain why?
I will listen some more to 6115 to see if things change for me.
stereophilus, Ted explained during beta testing that when the DS is powered off and then back on it takes a short while for the output transformers to… stabilize?.. get back into their own, for lack of a better way to put it. Should not take more than 10 minutes or so if it’s the normal quick power cycling done for a FW update.
As to the sound, I can only say (and believe me I hate to say) that it’s possible some re-evaluation of components (specifically cables) may be necessary. The ICs I was using with the PWD were terrible with the DS, and based on my memory of how I selected them the others I did try - like the Zitron Pythons - I estimate would have been horrible with the DS (at least in my system). I believe I’ve found a better choice for the DS, but it was clear a change was needed at least for me.
assisi said
I have had my DS upgrade for a week. 100 hour of running. Very impressed with it straight out of the box and it has continued to improve noticeably over time. Loaded the new firmware today. The new firmware is another step altogether. Female vocals are incredibly real. The air, life and musicality of the new sound is exquisite. Very very pleased!
tony22 said
stereophilus, Ted explained during beta testing that when the DS is powered off and then back on it takes a short while for the output transformers to... stabilize?... get back into their own, for lack of a better way to put it. Should not take more than 10 minutes or so if it's the normal quick power cycling done for a FW update.
Ok, that makes more sense than firmware break-in!
As to the sound, I can only say (and believe me I hate to say) that it's possible some re-evaluation of components (specifically cables) may be necessary. The ICs I was using with the PWD were terrible with the DS, and based on my memory of how I selected them the others I did try - like the Zitron Pythons - I estimate would have been horrible with the DS (at least in my system). I believe I've found a better choice for the DS, but it was clear a change was needed at least for me.
I might do more firmware listening before I rush to change my cables in what has been a very stable and rewarding system for me.
Ugh, I’m unable to get it to install, I’ve done it in the past no problems but not this time. I’ve tried both a 2 and 1 gb SD. I am certain the files were unzipped before copying to the SD, and have tried copying just the large file and both, ie the Mac (?) too. I removed old files beforehand. I’m using a win PC to extract.
Is a FAT formatted card ok ? What could be wrong ?
Paul mentions “letting it settle” when upgrading to the new firmware…? Is there an expected break-in for firmware?
not sure of the firmware naming convention - the new one is 6115 apparently, which seems to be a mash of CPLD 0.61 and version 1.1.5. So the original is CPLD 0.54 version 1.1.2 should be 5412.
I listened to 5412 for 30 minutes, then 6115 for an hour, then back to 5412 for 30 minutes. I prefer 5412. It has less edginess, it is more subtle, fluid and is more musical. 6115 sounds more like the PWD2 to me, with a harder, more pronounced leading edge on notes.
Perhaps I need to let 6115 settle for a longer period. If so, someone please explain why?
I will listen some more to 6115 to see if things change for me.
It could easily be your system doesn’t jive as well with the new software but I assure you, it’s better, more musical than the older. There’s really no settling in of firmware. I think it must be more the listener has to get adjusted, resettle the equipment, cabling choices and speaker placement to optimize your setup for 6115. It really is better in all respects. It’s rare we can just plunk something in and have it work without any further adjusting.
FAT is perfect. Users report success with both FAT16 and FAT32. I doubt it is a formatting issue if the card you are using otherwise worked.
The files need be in the root directory. That is, not in a folder on the SD card. An easy way to insure this is to delete everything on the card and then copy over only the six files comprising the firmware upgrade. These are the files that need to be on the SD card.
Turn of the DS, remove the SD card, delete the existing files on the card, copy over the new files, re-insert card, turn DS back on.
I might do more firmware listening before I rush to change my cables in what has been a very stable and rewarding system for me.
Always a good idea. I don’t recall if you said, but how many hours on your DS? If it hasn’t gotten into the hundreds of hours yet you may also be dealing with the combination of the increased BW in the new FW and (what some of us observed as) the rough edges in a couple of spots in the DS which smooths out after burn in. I think I can see a relationship between the two adding up in a way that you’ve described.
Elk said
FAT is perfect. Users report success with both FAT16 and FAT32. I doubt it is a formatting issue if the card you are using otherwise worked.
The files need be in the root directory. That is, not in a folder on the SD card. An easy way to insure this is to delete everything on the card and then copy over only the six files comprising the firmware upgrade. These are the files that need to be on the SD card.
Turn of the DS, remove the SD card, delete the existing files on the card, copy over the new files, re-insert card, turn DS back on.
I may be in the minority, but to my ears 6115 does not sound good in my system. That may be due to beryllium drivers supplemented by a Stat supertweeter, which deliver a tremendous amount of information and are phase coherent (at least from 250-8K) due to the concentric drivers.
What I am hearing is similar to how early LCD tv’s display noise and pixel creep in signals that wasn’t meant to be decoded. I am hearing digital sounding aura around voices, as if the detail control on a tv has been raised by a sharpening filter that increases apparent resolution. Could this possibly be dither?
It wasn’t until I went back to the original shipping kit version of the FW, and all was well. While I will say that I heard a bit more of the room ambience with 6115, I can’t say it sounded more real, only more there.
I’m not saying that the original firmwear is perfect and couldn’t possibly benefit from a bit more air, but it is of a whole, more musical and fully enjoyable.
I’m listening through a top tube preamp and DS at full volume with about 3-400 hours on the unit.
Interestingly, I took my DS to a friend that has the PWD. His system consists of a high level triamp system, capable of some of the most impressive dynamics I’ve heard in a long time. But he was using a interim series of speakers that were rather low end and while the dynamics were great, the resolution wasn’t there.
The difference between the PWD1 and DS (in that system/via optical) was not that pronounced, and though I was so familiar with what the DS did, I could see someone preferring the PWD’s balance.
In my system the DS is (at times) another world, so perhaps what we’re hearing is very system dependent.
adriaan said
@Ted or @Paul : How were you able to lower the jitter even further using the new firmware?
Tho the input jitter rejection is very good, just like with any reclocker, more jitter can creep in downstream. In this case the FPGA itself can generate jitter or noise that needs to be cleaned up downstream. I have a reclocker right at the clock proper, but I could have added more hardware to further lessen the effects of FPGA jitter. The balance between having to spend some time playing around in the FPGA vs. a higher price seemed to favor a little more pain on our part than in your pocket book. On a cost no object DAC I would have done it differently (and did for my prototype.)
Anyway I’ve been developing a “feel” for some of the simpler tricks to avoid generating more jitter in the FPGA. We will always have to check a particular compile with our ears, but I hope to make that less critical as time goes on.
Paul mentions “letting it settle” when upgrading to the new firmware…? Is there an expected break-in for firmware?
not sure of the firmware naming convention - the new one is 6115 apparently, which seems to be a mash of CPLD 0.61 and version 1.1.5. So the original is CPLD 0.54 version 1.1.2 should be 5412.
I listened to 5412 for 30 minutes, then 6115 for an hour, then back to 5412 for 30 minutes. I prefer 5412. It has less edginess, it is more subtle, fluid and is more musical. 6115 sounds more like the PWD2 to me, with a harder, more pronounced leading edge on notes.
Perhaps I need to let 6115 settle for a longer period. If so, someone please explain why?
I will listen some more to 6115 to see if things change for me.
The FPGA naming convention is simple: Increment the first two digits on any checked in feature change in the FPGA. We listen to about 20 compiles of each FPGA release: they are numbered, say 6101… 6120 for reference during testing. Each version differs in the seed for the random number generator in the compile process. (On more complicated FPGAs you may farm out many compiles across a sea of computers looking for just one that might complete successfully or the first one complete in a time crunch. The DS software is simple enough that it almost always compiles successfully, but still each compile makes different choices for gate assignment, routing, etc. which may affect noise generation or even jitter on the output lines.)
As others have mentioned some small amount of “settling” in time after power up is required, on the order of minutes not hours nor days. I just know that quick A/Bs are a complete waste for me and that others have reported things sounding better a little while after changing the software. I speculate that it’s the transformer reaching steady state (I know that sounds weird, but I’m not using the transformer with a pure audio signal…) I’ve also heard speculation that the clock might need time to reach steady state. I think this is less likely, but nevertheless it’s possible.
The other possibilities that people mention related to break-in are also plausible IMO. FWIW Tho, arguably, claiming longer required burn-in could work to my advantage, I still remain mildly skeptical about the length of burn ins that are being reported.
I just know that from a technical viewpoint in addition to some bug fixes 6115 should sound better, but YMMV
What I am hearing is similar to how early LCD tv's display noise and pixel creep in signals that wasn't meant to be decoded. I am hearing digital sounding aura around voices, as if the detail control on a tv has been raised by a sharpening filter that increases apparent resolution.
I have always appreciated this analogy. It is wonderfully descriptive.
Does the sound appear to be accompanied by a grittiness as well, or is the sound transparent but with sharp edges?
