Feyire
Active Member
Given that we have devices that improve in measured performance when warmed up over time and others that degrade, I'd really like to see this type of SINAD over time measurement became standard here at ASR.
-
WANTED: Happy members who like to discuss audio and other topics related to our interest. Desire to learn and share knowledge of science required. There are many reviews of audio hardware and expert members to help answer your questions. Click here to have your audio equipment measured for free!
SMSL M500 DAC Only “Thermal Issue” not “Golden Sample”
- Thread starter WolfX-700
- Start date
Disassembly?Why do people say "teardown"? It's such a brutal word and suggests irreversible damage, stripped screws and broken chit.
Carefully dismantle would be an easier sell to your friend.
- Joined
- Nov 6, 2018
- Messages
- 1,465
- Likes
- 4,889
Do we have DACs improving as they warm up? I am not well informed in the details of their low level physical implementations, but in other sensors (imaging) an increase in temperature always means an increase in dark current and ADC noise.
Feyire
Active Member
Feyire
Active Member
@amirm, so the hump is not present when cold, but appears when warm? If so, that opens up Pandora's box on the IMD hump issue once again... and puts into question whether previously measured devices without the hump only did so because they were cold/colder. For example, you would need to revisit ESS DACs, but also AKM DACs again.
This will continue to be a very interesting issue to investigate
This will continue to be a very interesting issue to investigate
How tf??? Thanks for measuring. I was thinking like, perhaps we get to know what component that changed performance. Turns out it's unknown again.... What about other dacs that didn't have imd hump shown?
Last edited:
I have run this with amplifiers. Here is an example of Hypex NC400:
I don't think it pays to worry about such things. I divide the DAC SINAD into just four buckets for this reason. A few dBs here and there don't mean anything in practice.
You guys are determined to glue me to my desk the entire weekend performing these tests.How tf??? Thanks for measuring. I was thinking like, perhaps we get to know what component that changed performance. Turns out it's unknown again.... What about other dacs that doesn't have imd hump shown?
I just packed everything I had into the attic to clean up my overflowing lab. So digging up old products takes a lot of effort and pain (can't stand up in the attic). I will see if I can find another instance to test.
Feyire
Active Member
You guys are determined to glue me to my desk the entire weekend performing these tests.
I just packed everything I had into the attic to clean up my overflowing lab. So digging up old products takes a lot of effort and pain (can't stand up in the attic). I will see if I can find another instance to test.
You said that Topping claimed they definitively fixed the hump issue on the D50S, which you confirmed with a measurement.
It would be really interesting to see a new set of measurements on that device with the SINAD over time chart as well as including IMD measurements over time to see if the hump re-appears at a later stage or not.
Testing D50s warm-up.... Report back in a couple of hours...
Feyire
Active Member
- Joined
- Feb 5, 2019
- Messages
- 3,699
- Likes
- 4,101
Can't you put it in the oven?
I can't wait.
- Joined
- Jan 20, 2019
- Messages
- 344
- Likes
- 501
Well my m500 has been on all day and sounds awesome right now, so for all practical purposes, these deviations are moot.
Regardless, thanks for all the efforts. Your work will definitely push the technology forward even further.
Regardless, thanks for all the efforts. Your work will definitely push the technology forward even further.
solderdude
Grand Contributor
Dang, I was always insured by goldeneared people stuff only improves on warm-up.
It will, just wait another week or two.
restorer-john
Grand Contributor
Whatever way you look at it, all testing going forward needs to be at stabilized operating temperature after an hour (graphs seem to show that would be sufficient).
Manufacturers will scream tests results are not typical of real-world usage otherwise.
Manufacturers will scream tests results are not typical of real-world usage otherwise.
Similar threads
- Poll
- Poll
