The No.1 Website for Pro Audio
ECC vs. non ECC memory
Old 10th February 2021 | Show parent
  #31
Here for the gear
 
🎧 5 years
Quote:
Originally Posted by uOpt ➡️
Supermicro has several variants of the X12SCZ board, which are mini-ATX Xeon ECC capable. Getting a LGA1200 Xeon in stock anywhere might be a different matter. But then all attractive CPUs and GPUs are sold out everywhere...

I agree that watching Blu Rays in your player software of choice is the way to go.

You might have more luck with in-storeness in the AMD world. You can buy an older CPU now and put in a 5000 series later.
I'm guessing the problem with buying Xeon server CPUs (as opposed to consumer versions) is that they're usually sold in lots, so even if I have Microcenter build the desktop while they'll have no problem getting the ECC compatible mobo getting a single CPU chip could take months? If yes, that's okay; no big rush, just as long as it happens.
Old 10th February 2021 | Show parent
  #32
Lives for gear
 
uOpt's Avatar
 
🎧 5 years
Quote:
Originally Posted by d1966 ➡️
I'm guessing the problem with buying Xeon server CPUs (as opposed to consumer versions) is that they're usually sold in lots, so even if I have Microcenter build the desktop while they'll have no problem getting the ECC compatible mobo getting a single CPU chip could take months? If yes, that's okay; no big rush, just as long as it happens.
Right now Newegg has the 6-core in stock:
https://www.newegg.com/intel-xeon-w-...-203-_-Product
Old 10th February 2021 | Show parent
  #33
Here for the gear
 
🎧 5 years
Quote:
Originally Posted by uOpt ➡️
Right now Newegg has the 6-core in stock:
https://www.newegg.com/intel-xeon-w-...-203-_-Product
Link doesn't seem to work.
Old 10th February 2021 | Show parent
  #34
Lives for gear
 
uOpt's Avatar
 
🎧 5 years
Quote:
Originally Posted by d1966 ➡️
Link doesn't seem to work.
Looks like a problem with the GS URL rewriter.

You can search on Newegg: Xeon in the text field, then CPUs, then LGA1200.
Old 10th February 2021 | Show parent
  #35
Here for the gear
 
🎧 5 years
Quote:
Originally Posted by uOpt ➡️
Looks like a problem with the GS URL rewriter.

You can search on Newegg: Xeon in the text field, then CPUs, then LGA1200.
Yes, I entered Xeon CPU LGA1200 into search at New Egg and there's mobos from Gigabyte, ASUS and AsRock. And one six core Comet Lake chip Xeon W-1250 # CM8070104379507. Presumably, this will all work great especially if large fans and heat pipes can make the pc as quiet as the fan in my Oppo BDP-95 player. Or if not Comet's Rocket Lake successor out next month then the 10nm based Alder Lake out later in the year provide that Oppo player level of quietude?
Old 10th February 2021 | Show parent
  #36
Lives for gear
 
uOpt's Avatar
 
🎧 5 years
Quote:
Originally Posted by d1966 ➡️
Yes, I entered Xeon CPU LGA1200 into search at New Egg and there's mobos from Gigabyte, ASUS and AsRock. And one six core Comet Lake chip Xeon W-1250 # CM8070104379507. Presumably, this will all work great especially if large fans and heat pipes can make the pc as quiet as the fan in my Oppo BDP-95 player. Or if not Comet's Rocket Lake successor out next month then the 10nm based Alder Lake out later in the year provide that Oppo player level of quietude?
I'm not sure this will be making much of a difference in heat to dissipate.

My recipe for a quit PC is a big steel (not alu) case with several 120mm exhaust fans. You want to move a lot of air slowly, not less air quickly.

Of course a fanless BR player is hard to beat noise-wise
Old 10th February 2021
  #37
Here for the gear
 
🎧 5 years
Quote:
Originally Posted by uOpt ➡️
Quote:
Originally Posted by d1966 ➡️
Yes, I entered Xeon CPU LGA1200 into search at New Egg and there's mobos from Gigabyte, ASUS and AsRock. And one six core Comet Lake chip Xeon W-1250 # CM8070104379507. Presumably, this will all work great especially if large fans and heat pipes can make the pc as quiet as the fan in my Oppo BDP-95 player. Or if not Comet's Rocket Lake successor out next month then the 10nm based Alder Lake out later in the year provide that Oppo player level of quietude?
I'm not sure this will be making much of a difference in heat to dissipate.

My recipe for a quit PC is a big steel (not alu) case with several 120mm exhaust fans. You want to move a lot of air slowly, not less air quickly.

Of course a fanless BR player is hard to beat noise-wise
For 1080p (never any 4K) video editing using this software https://www.cyberlink.com/products/v...ecs_en_US.html but doing no more than deleting scenes and/or attenuating loud noises-wouldn’t a six core Xeon CPU, a healthy amount of ECC memory and a good but not beasty graphics card make quick work of such jobs while hardly breaking a sweat to create much fan noise, if only just very briefly?

As for using the desktop for actual entertainment, I’m not the least bit into gaming nor watch many movies from the Action genre. So except for maybe something like an old James Bond movie from the early 60s (or less often the Girl with the Dragon Tattoo), almost none of the two hour BD movies I’d be watching on the desktop via JRiver player would be pushing the graphics card to do lengthily renderings of rapidly changing images at 1080p.

Given those tasks, hardware and software, would you still expect the radiator cooled cpu and gpu with two or three large slowly spinning 120mm fans in a “steel” mini-ATX case to make appreciable heat? And noticeable fan noise not masked by a 5.1 system playing at ~ 45db in a ~ 14 x 18 ft room?

Furthermore, while I’m sure about this, JRiver and/or the video card’s software interface may be settable where when playing a 1080p BD movie disc to be displayed on an OLED TV-all of which are 4K-the pc outputs non-upscaled video to the TV. And the TV can be set to do the upscaling thereby no chance of more heat and fan noise from the pc.

As for the next gen 10nm architecture, I thought one of its biggest advantages was efficiency, which would mean doing all of the above work with less heat and so less fan noise.
Old 12th February 2021
  #38
Here for the gear
 
🎧 5 years
Quote:
Originally Posted by MediaGary ➡️
Quote:
Originally Posted by Alex Dunnett ➡️
I was reading that ECC has around 33% less bit slip somewhere.. ..
I'd like to read that reference. The 'bit slip' term usually arises in the context of serial data communications where the synch between transmitter and receiver isn't stable.
As uOpt mentioned, storage HDDs have substantial error correction, but except perhaps only for Xeon server or comparable AMD cpu-supporting motherboards, most pcs have motherboards which do not support ECC memory.

Thus, during single CD track ripping/save to file and routine backup sessions, is there software or hardware (obviously designed to operate independently of anything resembling non-ECC memory), which can count the number of bit flips that may have happened while a particular CD track was being ripped and saved (to an uncompressed WAV file) or a file was being copied or moved from one internal or external storage device to another within a pc’s system?

Could it also do things like indicate which particular file the bit flip had occurred while a dozen or so files were being simultaneously copied or moved to another drive?

If yes, who makes among the most accurate and easy to use solutions?
Old 12th February 2021 | Show parent
  #39
Lives for gear
 
uOpt's Avatar
 
🎧 5 years
Quote:
Originally Posted by d1966 ➡️
As uOpt mentioned, storage HDDs have substantial error correction, but except perhaps only for Xeon server or comparable AMD cpu-supporting motherboards, most pcs have motherboards which do not support ECC memory.
AMD doesn't need anything special if you use any of the GS recommended boards.

Quote:
Thus, during single CD track ripping/save to file and routine backup sessions, is there software or hardware (obviously designed to operate independently of anything resembling non-ECC memory), which can count the number of bit flips that may have happened while a particular CD track was being ripped and saved (to an uncompressed WAV file) or a file was being copied or moved from one internal or external storage device to another within a pc’s system?

Could it also do things like indicate which particular file the bit flip had occurred while a dozen or so files were being simultaneously copied or moved to another drive?

If yes, who makes among the most accurate and easy to use solutions?
The cannonical solution for this scenario is to sample 3 times, preferably on different systems, and compare the resulting files.

For me, the major concern always is a valuable file such as a recording being corrupted. Possibly silently so, and possibly long after you intentionally wrote it.

And I am not so afraid of random bit flips from cosmic rays. For me it is more about RAM malfunctioning suddenly such as from a heat problem, or RAM chips going bad. If you run like that for a while that is very toxic for the filesystem. And without ECC you lack both the protection and the reporting.
Old 12th February 2021 | Show parent
  #40
Here for the gear
 
🎧 5 years
Quote:
Originally Posted by uOpt ➡️
AMD doesn't need anything special if you use any of the GS recommended boards.



The cannonical solution for this scenario is to sample 3 times, preferably on different systems, and compare the resulting files.

For me, the major concern always is a valuable file such as a recording being corrupted. Possibly silently so, and possibly long after you intentionally wrote it.

And I am not so afraid of random bit flips from cosmic rays. For me it is more about RAM malfunctioning suddenly such as from a heat problem, or RAM chips going bad. If you run like that for a while that is very toxic for the filesystem. And without ECC you lack both the protection and the reporting.
I'm sold. Two custom built Xeon ECC memory desktops for me this year. Thanks so much for educating me on how critical ECC memory is for ensuring bit integrity for audio files. The audiophile community were done a huge disservice by the underplaying of this serious reality. And much of the blame looks due to decades of Intel's mass marketing of cpus/motherboards which were not ECC supportive. Had they simply sold Xeon processors to both consumer and server markets even if ECC memory was slightly more costly economies of scale would have quickly eclipsed that difference.
Old 13th February 2021 | Show parent
  #41
Here for the gear
 
🎧 5 years
Quote:
Originally Posted by uOpt ➡️
For me, the major concern always is a valuable file such as a recording being corrupted. Possibly silently so, and possibly long after you intentionally wrote it.

And I am not so afraid of random bit flips from cosmic rays. For me it is more about RAM malfunctioning suddenly such as from a heat problem, or RAM chips going bad. If you run like that for a while that is very toxic for the filesystem. And without ECC you lack both the protection and the reporting.
Better yet, one ultra quiet Xeon/ECC desktop for web surfing, downloading and word processing one desktop for redundant backup and with plenty of horsepower for 1080p video editing.

Important question: Suppose I’m downloading a youtube on this laptop, later backing up that file to a storage drive on this desktop, ripping a CD track and/or saving it to a WAV file on this desktop. While doing so ECC memory detects a bit error. Does it then flash a message that this happened? Does it also say that it either successfully corrected the error or not? Or does it just flash the bit error message and then it’s to me “correct” the error? But how else could I correct the error other than deleting the file and creating a new one by copying the original or re-rip and save the CD track to a new file?
Old 13th February 2021 | Show parent
  #42
Lives for gear
 
uOpt's Avatar
 
🎧 5 years
Quote:
Originally Posted by d1966 ➡️
Important question: Suppose I’m downloading a youtube on this laptop, later backing up that file to a storage drive on this desktop, ripping a CD track and/or saving it to a WAV file on this desktop. While doing so ECC memory detects a bit error. Does it then flash a message that this happened? Does it also say that it either successfully corrected the error or not? Or does it just flash the bit error message and then it’s to me “correct” the error? But how else could I correct the error other than deleting the file and creating a new one by copying the original or re-rip and save the CD track to a new file?
You need to actively look at the system logs. A good idea anyway for other reasons, too.

1-bit errors are always reported and corrected.

2-bit errors are detected and not corrected. What the machines does with it depends on opinions, IMHO the machine should halt (like a bluescreen).
Old 13th February 2021 | Show parent
  #43
Gear Maniac
 
🎧 5 years
Also, regardless of what RAM you have (but especially with non-ECC RAM that runs at faster speeds) you should test the RAM for stability.
Don't assume an XMP profile is safe and just works 100% error free, that's often not the case.

My current favorite RAM test is MemTest5 with the "Extreme1 anta777" config--I found that to be the most thorough (run it for at least 8h by increasing the cycles). But there are others too and it's not a bad idea to run several different ones.
Old 15th February 2021 | Show parent
  #44
Here for the gear
 
🎧 5 years
Quote:
Originally Posted by uOpt ➡️
You need to actively look at the system logs. A good idea anyway for other reasons, too.

1-bit errors are always reported and corrected.

2-bit errors are detected and not corrected. What the machines does with it depends on opinions, IMHO the machine should halt (like a bluescreen).
This is crazy LOL embarrassing because I’ve never once looked at system error logs. It never occurred to me that the cause or at least the occurrence of audio related errors, if they were noted by the system at all, would (hopefully) be intelligible to the less rigorously trained Windows users. Instead, along with temp files I’ve long been deleting them during periodic disc cleanup sessions!! And as I rip CD tracks less than frequently there are likely no relevant error logs to look at currently, if such errors were created. But I certainly will Uncheck the system error file box from disc cleanup and keep it that way.

BUT do you need to be a coder to read those error log files and then to make use of the info to correct the problem and hopefully prevent such future problems? Beyond memorizing a few DOS keystrokes I never learned how to do anything in the command prompt.

Quote:
Originally Posted by BenSW ➡️
Also, regardless of what RAM you have (but especially with non-ECC RAM that runs at faster speeds) you should test the RAM for stability. Don't assume an XMP profile is safe and just works 100% error free, that's often not the case.

My current favorite RAM test is MemTest5 with the "Extreme1 anta777" config--I found that to be the most thorough (run it for at least 8h by increasing the cycles). But there are others too and it's not a bad idea to run several different ones.
WOW two great error detecting/correcting/preventing suggestions in a row!

Yes, I think I remember from tech school long ago that non-ECC’s advantage is that it’s faster than ECC RAM since it doesn’t check for errors (using parity bit checking). Perhaps that’s why since the speed thirsty warfare, gaming (and some have suggested porn) markets drive the computer market, non-ECC memory became more prevalent, whereas mission critical error intolerant apps in R & D (medicine, next gen energy and pure science) which need ECC memory are generally lower volume markets.

I’ve seen posts from advanced gamers and others who over clock their pcs discussing their RAM testing results, but I never thought that would be required for those into high resolution audio, or even 1080p video editing (the latter I won’t even be doing much of). But as I mentioned here that I’ve zero interest in gaming or any kind of math intensive computing what makes you think that I’d be looking to (have Micro Center) over clock my new hardware? https://forums.tomshardware.com/thre...stion.3638559/ https://www.overclock.net/threads/ru...c-mode.851193/

I’m too ignorant to even follow the preceding discussions. Thus, I would hardly be setting up any XMP profiles in the BIOS.

Kode54 here, reply #11 , https://hydrogenaud.io/index.php?top...2985#msg922985 suggests that ECC memory is more prone to crashes and creating other errors. But would the chances of that happening be far less likely in pcs that did zero gaming or other CPU intensive work, rather than any kind of high res stereo or multichannel audio home theater apps?

And why would I need to over clock a six core Xeon Rocket Lake system even if not running the faster non-ECC RAM, assuming it’s safe and cost effective to even over clock ECC RAM? Doing 1080p (never 4K) BD movie decrypting, a few simple scene deletions and a finally a BD-R burn-much less editing 16 or 24 bit audio-would be tasks hardly calling for an over clocked pc.

But as you say, any kind of RAM needs to be periodically tested, though as you can see from my last few posts I’m no kind of command prompt user!

Therefore, please recommend the most thorough but easiest to use (Windows based??) RAM chip integrity software (and hardware?).

And how often should I run those RAM tests on the ~ 8 to 16GB of ECC RAM I’d have in something like a Xeon Rocket Lake cpu supporting motherboard? Once a month?

Another LOL moment: I have one Pentium IV 3.2GHz XP SP2 with 2GB non-ECC RAM running XP SP2. For now that’s part of my redundant backup chain. Can I run this same RAM integrity program on that dinosaur too? If yes, any special precautions?

Also that the consensus here
https://hydrogenaud.io/index.php?topic=109544.0
https://hydrogenaud.io/index.php?topic=109544.25
https://hydrogenaud.io/index.php?topic=109544.25
is that saving CD track rips to FLAC files is better than doing so to WAV as FLAC does check sum error detection. But it is silly to attempt offering a crash course in explaining check sum error detection, where to access that info and what to look which may indicate errors?

But to bring things full circle, on the one hand many have claimed that when bit errors do impact audio files the effects are almost always extreme. For example, what happened to this guy’s FLAC files. https://jens-ingo.all2all.org/archives/1900

But I must have ripped at least 80 CD tracks to uncompressed WAV files and I’ve never experienced the kinds of errors he had.

On the other hand, are there other kinds of audio quality losses that can occur due to bit flips? For example, while the DAC in my old ASUS sound card may limit my aural detection of this, how likely can bit errors not cause crashes or other big problems but can cause gradual or even sudden losses in audio resolution and other even barely audible damage in WAV or FLAC files which likely be noticeable with a better DAC?

And while possible with any RAM, can resolution loss be more likely to happen in non-ECC RAM pcs?

Also, whether saving rips to WAV or FLAC, as Rosval suggests here https://hydrogenaud.io/index.php?top...ew#info_993800 there will be no chance of any data corruption if the file is saved as read only?
Old 15th February 2021 | Show parent
  #45
Gear Maniac
 
🎧 5 years
Quote:
Originally Posted by d1966 ➡️
And how often should I run those RAM tests on the ~ 8 to 16GB of ECC RAM I’d have in something like a Xeon Rocket Lake cpu supporting motherboard? Once a month?
You definitely don't need to test once a month. Typically you'd test it once and if it passes then you're good (unless you start noticing something weird with your PC). But test it thoroughly, spend a couple of days and run a couple of different tests. I recommend HCI Memtest and TestMem5. You'll find more about those tests on various overclocking forums.
Mind you, I don't think it's very likely that you'll get errors with standard speed RAM (say 2666 MHz DDR4). It's a bit more of a concern with 3200+ MHz RAM with tight timings.

Also, regarding ECC RAM speed: it's not that it's significantly slower at the same MHz and timings. IIRC the ECC speed penalty is negligible, something like 1%.
But the thing is that they just don't sell ECC RAM that runs at very high speeds. You can buy non-ECC DDR4 RAM nowadays that runs at 4000 MHz CL16, but the ECC RAM sold is typically under 3000 Mhz. There is some ECC RAM at 3200 MHz, but it's CL22 (higher CL = slower).
How much this matters is up to you. The RAM speed impact is probably a bit overrated, but it's better if you just look at the actual benchmarks and decide for yourself.
Old 15th February 2021 | Show parent
  #46
Lives for gear
 
uOpt's Avatar
 
🎧 5 years
Quote:
Originally Posted by d1966 ➡️
BUT do you need to be a coder to read those error log files and then to make use of the info to correct the problem and hopefully prevent such future problems? Beyond memorizing a few DOS keystrokes I never learned how to do anything in the command prompt.
The ECC error messages should be pretty readable.

Quote:
Kode54 here, reply #11 , https://hydrogenaud.io/index.php?top...2985#msg922985 suggests that ECC memory is more prone to crashes and creating other errors.
Well, that applies if the system is halted (or bluescreened) when 2-bit errors happen. Of course you wouldn't have the system halted if you couldn't detect the errors in the first place.

But you really don't want to continue to run a system that has acute 2-bit errors.

Quote:
And why would I need to over clock a six core Xeon Rocket Lake system even if not running the faster non-ECC RAM, assuming it’s safe and cost effective to even over clock ECC RAM? Doing 1080p (never 4K) BD movie decrypting, a few simple scene deletions and a finally a BD-R burn-much less editing 16 or 24 bit audio-would be tasks hardly calling for an over clocked pc.
CPU speed is much more important than RAM speed. Just because you don't overclock your RAM doesn't mean that overclocking your CPU is useless.

Quote:
But as you say, any kind of RAM needs to be periodically tested, though as you can see from my last few posts I’m no kind of command prompt user!
If you have ECC memory you can test while the system is operating. It is called "scrub". You turn it on in the BIOS and it goes through all the RAM periodically to see whether any ECC errors are being triggered.

Now, unlike just running ECC plain the scrubbing comes with a performance penalty.

Quote:
Another LOL moment: I have one Pentium IV 3.2GHz XP SP2 with 2GB non-ECC RAM running XP SP2. For now that’s part of my redundant backup chain. Can I run this same RAM integrity program on that dinosaur too? If yes, any special precautions?

Also that the consensus here
https://hydrogenaud.io/index.php?topic=109544.0
https://hydrogenaud.io/index.php?topic=109544.25
https://hydrogenaud.io/index.php?topic=109544.25
is that saving CD track rips to FLAC files is better than doing so to WAV as FLAC does check sum error detection. But it is silly to attempt offering a crash course in explaining check sum error detection, where to access that info and what to look which may indicate errors?

But to bring things full circle, on the one hand many have claimed that when bit errors do impact audio files the effects are almost always extreme. For example, what happened to this guy’s FLAC files. https://jens-ingo.all2all.org/archives/1900

But I must have ripped at least 80 CD tracks to uncompressed WAV files and I’ve never experienced the kinds of errors he had.

On the other hand, are there other kinds of audio quality losses that can occur due to bit flips? For example, while the DAC in my old ASUS sound card may limit my aural detection of this, how likely can bit errors not cause crashes or other big problems but can cause gradual or even sudden losses in audio resolution and other even barely audible damage in WAV or FLAC files which likely be noticeable with a better DAC?

And while possible with any RAM, can resolution loss be more likely to happen in non-ECC RAM pcs?

Also, whether saving rips to WAV or FLAC, as Rosval suggests here https://hydrogenaud.io/index.php?top...ew#info_993800 there will be no chance of any data corruption if the file is saved as read only?
A lot to unpack here.

Saving a file as readonly is useless in our context here. Corruption can hit it the same way. It's not that you magically have a readonly storage medium.

Harddrives have more than just a checksum (CRC), they rarely flip bit the way non-ECC RAM can.

I support the notion that going from ripping to FLAC in one program run improves reliability.

If you really want best results you might want to consider running a filesystem that has extra integrity checking such as ZFS. But we are leaving the zone of Windows here.
Old 15th February 2021 | Show parent
  #47
Gear Addict
 
snoskit's Avatar
 
Quote:
Originally Posted by uOpt ➡️
...

If you really want best results you might want to consider running a filesystem that has extra integrity checking such as ZFS. But we are leaving the zone of Windows here.
Windows has ReFS, which ties in with StorageSpaces.
Old 15th February 2021 | Show parent
  #48
Gear Addict
 
snoskit's Avatar
 
Quote:
Originally Posted by d1966 ➡️
But to bring things full circle, on the one hand many have claimed that when bit errors do impact audio files the effects are almost always extreme.
IME corruption effects compressed files more adversely than uncompressed data.
FLAC, ZIP, RAR, etc
Compressed NTFS volumes too.

You can create checksum files (Certutil/get-filehash) for waves files.
Old 16th February 2021 | Show parent
  #49
Here for the gear
 
🎧 5 years
Quote:
Originally Posted by BenSW ➡️
Also, regarding ECC RAM speed: it's not that it's significantly slower at the same MHz and timings. IIRC the ECC speed penalty is negligible, something like 1%. But the thing is that they just don't sell ECC RAM that runs at very high speeds. You can buy non-ECC DDR4 RAM nowadays that runs at 4000 MHz CL16, but the ECC RAM sold is typically under 3000 Mhz. There is some ECC RAM at 3200 MHz, but it's CL22 (higher CL = slower). How much this matters is up to you. The RAM speed impact is probably a bit overrated, but it's better if you just look at the actual benchmarks and decide for yourself.
Again, the most demanding thing I’d ever ask of this new build would be to fairly often play two hour 1080p BD movies with this https://www.pioneerelectronics.com/P...Drives.Support and this https://jriver.com/overview.html

Or only occasionally use software like this to rip BDs (no 4K) or DVDs https://www.redfox.bz/en/anydvdhd.html and then do simple frame or scene deletions and attenuations using this https://www.cyberlink.com/products/p...ecs_en_US.html and finally burning the finished job to a BD-R or BD-RE.

So how fast should the ECC RAM that I buy be?
Old 16th February 2021 | Show parent
  #50
Here for the gear
 
🎧 5 years
Quote:
Originally Posted by uOpt ➡️
The ECC error messages should be pretty readable.
Good news for a change. Thanks.

Quote:
Originally Posted by uOpt ➡️
CPU speed is much more important than RAM speed. Just because you don't overclock your RAM doesn't mean that overclocking your CPU is useless.
Hardly useless but as mentioned above, since this new build will be central to my home theater it has run very quietly. But how quiet could Microcenter make a desktop with an overclocked CPU?

Quote:
Originally Posted by uOpt ➡️
If you really want best results you might want to consider running a filesystem that has extra integrity checking such as ZFS. But we are leaving the zone of Windows here.
Yes, what Porcus, replies 7 and 10, has been after me to implement; yeah right, as if I’d know how. https://hydrogenaud.io/index.php?top...3887#msg993887

Quote:
Originally Posted by uOpt ➡️
I support the notion that going from ripping to FLAC in one program run improves reliability.
Speaking of FLAC vs WAV, as WAV doesn’t support error checking, like Porcus and most here https://hydrogenaud.io/index.php?topic=109544.0 https://hydrogenaud.io/index.php?topic=109544.25 you apparently therefore also favor CD track ripping directly to FLAC. While that’s clearly a sensible choice, based on what I read at two other forums today, please share you thoughts on Steve Nugent’s strange claims about “sound differences” between WAV and FLAC; reply # 64. https://forums.stevehoffman.tv/threa....978399/page-3

Could this somehow be due to any inherent compression in the otherwise lossless FLAC format, as opposed to the uncompressed WAV?
Old 16th February 2021 | Show parent
  #51
Here for the gear
 
🎧 5 years
Quote:
Originally Posted by snoskit ➡️
IME corruption effects compressed files more adversely than uncompressed data.
FLAC, ZIP, RAR, etc
Compressed NTFS volumes too.

You can create checksum files (Certutil/get-filehash) for waves files.
Ironic how a FLAC file would be more prone to corruption than a WAV file, the latter many here reject for its lack of error detection. https://hydrogenaud.io/index.php?topic=109544.0 https://hydrogenaud.io/index.php?topic=109544.25
Old 16th February 2021
  #52
Gear Maniac
 
🎧 5 years
@ d1966 : You're overwhelming yourself with questions that are barely related (and also off-topic in this thread). I suggest you take a step back and tackle one problem at a time, until you have a decent understanding of it; RAM, HDD errors, CD ripping, FLAC, computer noise... Pretty much all of this stuff has been discussed to death and answered many times on the web.
In terms of purchasing advice for a new PC, again, there are plenty of benchmarks and reviews detailing performance out there. But it sounds like you don't have any special requirements, so I think that pretty much any computer made in the last 5 years will be fast enough.
Old 16th February 2021 | Show parent
  #53
Lives for gear
 
uOpt's Avatar
 
🎧 5 years
Quote:
Originally Posted by d1966 ➡️
Speaking of FLAC vs WAV, as WAV doesn’t support error checking, like Porcus and most here https://hydrogenaud.io/index.php?topic=109544.0 https://hydrogenaud.io/index.php?topic=109544.25 you apparently therefore also favor CD track ripping directly to FLAC. While that’s clearly a sensible choice, based on what I read at two other forums today, please share you thoughts on Steve Nugent’s strange claims about “sound differences” between WAV and FLAC; reply # 64. https://forums.stevehoffman.tv/threa....978399/page-3

Could this somehow be due to any inherent compression in the otherwise lossless FLAC format, as opposed to the uncompressed WAV?
No, that is a misquote (apart from being obvious nonsense). Even the supposed original author says here that flac converted back to wav is identical to wav:

https://www.whatsbestforum.com/threa...24/post-537969
📝 Reply

Similar Threads

Thread / Thread Starter Replies / Views Last Post
replies: 0 views: 4248
Avatar for Henrik
Henrik 7th March 2003
replies: 14181 views: 1587019
Avatar for makeanoise
makeanoise 6 hours ago
Post Reply

Welcome to the Gearspace Pro Audio Community!

Registration benefits include:
  • The ability to reply to and create new discussions
  • Access to members-only giveaways & competitions
  • Interact with VIP industry experts in our guest Q&As
  • Access to members-only sub forum discussions
  • Access to members-only Chat Room
  • Get INSTANT ACCESS to the world's best private pro audio Classifieds for only USD $20/year
  • Promote your eBay auctions and Reverb.com listings for free
  • Remove this message!
You need an account to post a reply. Create a username and password below and an account will be created and your post entered.


 
 
Slide to join now Processing…

Forum Jump
Forum Jump