Perhaps you should try it for yourself.
Last edited:
![[H]ard|Forum](/styles/hardforum/xenforo/logo_dark.png){kind=logo}
-
Some users have recently had their accounts hijacked. It seems that the now defunct EVGA forums might have compromised your password there and seems many are using the same PW here. We would suggest you UPDATE YOUR PASSWORD and TURN ON 2FA for your account here to further secure it. None of the compromised accounts had 2FA turned on.
Once you have enabled 2FA, your account will be updated soon to show a badge, letting other members know that you use 2FA to protect your account. This should be beneficial for everyone that uses FSFT.
cables... finally convinced
- Thread starter CrazyCuz2k
- Start date
Perhaps you should try it for yourself.
minus 15% restocking fee of courseIf I am not blown away by them will you refund my money?
Fostex sends 300 unit shipments every few months, so they keep going in and out of stock. Maybe they heard them stock or without a good mod job.
I believe the T50RP modded is the best closed can available, though I haven't heard the Ed. 8s which some say are amazing. Last Fall I went to a meet and heard LFF's Paradox T50RPs, MrSpeaker's Rastapants2 T50RPs, Stax SR-009/BHSE, HD800/Balancing Act, and LCD-2/Lyr. Plenty of T50RP mods mixed with $10,000+ rigs
. My Fostex's and O2 are getting 50/50 headtime with my HD800s and WA2. The HD800s have more detail and soundtage, the Fostex's have more speed (and Led Zepplin loves them). LFF has stated that his Paradox modded T50RPs with an O2 are his endgame rig. Here are a few quotes from the meet impressions thread: "Modded Fostex T50RP sounded stupendous. Great bass presence and solidity to sound all over the ranges. Midrange was not veiled or dull and was detailed enough out of several amps." ~ Jon L
"LFF's modded T50RPs: Highlight of the meet for me personally. Best ortho I've ever heard, only ortho that has a good amount of air and a nice soundstage. This sounded more like an open headphone really...it totally caught me off guard. Tonal balance was fantastic, very similar to the SR009 but wasn't as refined/fast/open." ~ n3rdling
My Fostex T40RP's are pictured with my WA2. I didn't get HD800's until after I heard purrin's set with his Balancing Act.
http://www.head-fi.org/t/577060/irvine-orange-county-meet-impressions
timestretch
Limp Gawd
- Joined
- Mar 19, 2011
- Messages
- 200
I've heard modded Fostexes, with Sony XB700 pads (and some other changes as well, I think? the owner was a very DiY kind of guy).
These particular modded Fostexes had some powerful and deep bass. However, the rest of the range seemed muffled. Perhaps I'd need my ears to get accustomed to it. I didn't like them much, though.
I would be interested in hearing more modded fostexes. Apparently they have great potential. I am a bit doubtful, though.
These particular modded Fostexes had some powerful and deep bass. However, the rest of the range seemed muffled. Perhaps I'd need my ears to get accustomed to it. I didn't like them much, though.
I would be interested in hearing more modded fostexes. Apparently they have great potential. I am a bit doubtful, though.
Some people don't like blowjobs and prefer handies. To each their own. Everyone hears differently.
Now the pads used seem to make a difference in sound. Those sony pads could indeed blow up the bass as they were designed for a BOOMIN headphone in the first place. Anyway, I am going to get a pair to try for myself. I want to get a pair of stock and modded to compare if I can afford it at the time.
Some folks on headfi, myself included, can be a bit over the OMG TOP at times. However, when several folks who have experience with many models of headphones, agree, well, there must be something there. That is what I look for usually. A person who has experience and does call bullshit when it is, gets my attention.
Now the pads used seem to make a difference in sound. Those sony pads could indeed blow up the bass as they were designed for a BOOMIN headphone in the first place. Anyway, I am going to get a pair to try for myself. I want to get a pair of stock and modded to compare if I can afford it at the time.
Some folks on headfi, myself included, can be a bit over the OMG TOP at times. However, when several folks who have experience with many models of headphones, agree, well, there must be something there. That is what I look for usually. A person who has experience and does call bullshit when it is, gets my attention.
HammerSandwich
[H]ard|Gawd
- Joined
- Nov 18, 2004
- Messages
- 1,126
I feel I wasted my time by posting.
minus 15% restocking fee of course
Maybe they heard them stock or without a good mod job.
I believe the T50RP modded is the best closed can available, though I haven't heard the Ed. 8s which some say are amazing. Last Fall I went to a meet and heard LFF's Paradox T50RPs, MrSpeaker's Rastapants2 T50RPs, Stax SR-009/BHSE, HD800/Balancing Act, and LCD-2/Lyr. Plenty of T50RP mods mixed with $10,000+ rigs
"Modded Fostex T50RP sounded stupendous. Great bass presence and solidity to sound all over the ranges. Midrange was not veiled or dull and was detailed enough out of several amps." ~ Jon L
"LFF's modded T50RPs: Highlight of the meet for me personally. Best ortho I've ever heard, only ortho that has a good amount of air and a nice soundstage. This sounded more like an open headphone really...it totally caught me off guard. Tonal balance was fantastic, very similar to the SR009 but wasn't as refined/fast/open." ~ n3rdling
My Fostex T40RP's are pictured with my WA2. I didn't get HD800's until after I heard purrin's set with his Balancing Act.
http://www.head-fi.org/t/577060/irvine-orange-county-meet-impressions
OK, you have me convinced. Now just to source a pair. I'll phone some local musical instrument stores and see if they have any for sale. Musical instrument stores often carry Fostex instead of brands like Sennheiser and beyerdynamic. I already have an extra set of Shure 840 pads to mod them with too. Think I will go with the cotton ball mod with a rim of plasticine around the drivers.
I read the article so it was not a waste of your time. I just don't give pro audio mags much credibility because their no. 1 job is to help sell advertiser's products and not tell the truth to the consumer.
I think it's pretty safe to say that everyone hears things differently. Some people actually do have better hearing than others. If you can't hear any difference between 320kbit MP3 and lossless, then that's a limitation of the output device, or (more likely) a limitiation of your ability to hear things.
wonderfield
Supreme [H]ardness
- Joined
- Dec 11, 2011
- Messages
- 7,396
You are right not to give Stereophile much credibility.
I haven't seen that 99% number before. I imagine it's probably closer to 90%, but that's really only a guess.
timestretch
Limp Gawd
- Joined
- Mar 19, 2011
- Messages
- 200
On the subject of ABX, I recently did an ABX test in Foobar with an old song by the Cranberries (perhaps not the best choice; chose it randomly from my library - but it was a nice song so that's cool and it should work as well as most) and I scored 6/12 correct on determining flac vs 320kbps mp3 over the course of a 23 minute testing session.
I feel like I felt more confident before I guessed correctly, and whenever I was having trouble telling the difference I would get it wrong. If this has any basis, then perhaps the ability to hear the very subtle nuances between flac and 320bps is as much based on the immediate mood and mental state of the listener as anything else.
Feelings of pressure, or fatigue, or anything else, really, could distract you from the telling, differing details that you might otherwise hear if you were listening in "peak condition." In any case, the test was extremely difficult and I also think it's very probable that I was guessing the entire time. If there are audible differences, then they are *extremely* subtle.
I'll be trying more ABX tests in the future with a wider variety of recordings, on my current equipment and also on the new HE-6 I've recently ordered. I am hopeful that I will be able to pass the test eventually, and not because of pure luck, either. Maybe with denser recordings with more things going on, and higher dynamic range, I will be able to tell the two formats apart more easily. The Cranberries I suppose are not lauded for their magically true-to-life high-fidelity recordings.
But I am not terribly confident. 320kbps and flac are extremely similar. I would guess that most (if not very nearly all) people who say otherwise truly haven't compared them level matched and side to side, blind to the bit-rate.
I feel like I felt more confident before I guessed correctly, and whenever I was having trouble telling the difference I would get it wrong. If this has any basis, then perhaps the ability to hear the very subtle nuances between flac and 320bps is as much based on the immediate mood and mental state of the listener as anything else.
Feelings of pressure, or fatigue, or anything else, really, could distract you from the telling, differing details that you might otherwise hear if you were listening in "peak condition." In any case, the test was extremely difficult and I also think it's very probable that I was guessing the entire time. If there are audible differences, then they are *extremely* subtle.
I'll be trying more ABX tests in the future with a wider variety of recordings, on my current equipment and also on the new HE-6 I've recently ordered. I am hopeful that I will be able to pass the test eventually, and not because of pure luck, either. Maybe with denser recordings with more things going on, and higher dynamic range, I will be able to tell the two formats apart more easily. The Cranberries I suppose are not lauded for their magically true-to-life high-fidelity recordings.
But I am not terribly confident. 320kbps and flac are extremely similar. I would guess that most (if not very nearly all) people who say otherwise truly haven't compared them level matched and side to side, blind to the bit-rate.
When it comes to lossy versus lossless, it is also important to note that compression algorithms do not maintain the same efficiency across all content.
A recording of a violin, for instance, may be more difficult to compress than a recording of a bass guitar.
Let's say, for the sake of argument, that a bass guitar recording requires 300kbps worth of samples to be lossless. A violin recording, meanwhile, may require 800kbps worth of samples to be considered lossless. Here we are assuming that the bass guitar recording has more repetition and less dynamism than the recording of the violin.
Remember, compression works by taking advantage of repetition. So if an instrument produces and holds the exact same frequency for ten seconds, then we can sample the frequency ten times, once per second, or we can sample the frequency twice, once at second 1 and once at second 10, and extrapolate the rest. This is an oversimplification, but this is the basis of compression.
As such, if we assume a 320kbps encode, it would be possible to record the above bass guitar losslessly, even if the overall format is lossy, since we only need 300kbps of the allotted 320kbps to store all of the required samples.
Now, this isn't exactly how MP3 encoding works, but it is important to note that some recordings may very well be indistinguishable in the 320kbps format, while others will present readily apparent differences.
In my own library, there is plenty of music where I most likely would not be able to differentiate 320kbps MP3 from FLAC. At the same time, there exists music in my library, particularly symphonies, where FLAC is quite clearly superior. Simply put, if you have a symphony recording featuring 80 instruments, there may be too much going on for compression algorithms to correctly extrapolate the full experience.
A recording of a violin, for instance, may be more difficult to compress than a recording of a bass guitar.
Let's say, for the sake of argument, that a bass guitar recording requires 300kbps worth of samples to be lossless. A violin recording, meanwhile, may require 800kbps worth of samples to be considered lossless. Here we are assuming that the bass guitar recording has more repetition and less dynamism than the recording of the violin.
Remember, compression works by taking advantage of repetition. So if an instrument produces and holds the exact same frequency for ten seconds, then we can sample the frequency ten times, once per second, or we can sample the frequency twice, once at second 1 and once at second 10, and extrapolate the rest. This is an oversimplification, but this is the basis of compression.
As such, if we assume a 320kbps encode, it would be possible to record the above bass guitar losslessly, even if the overall format is lossy, since we only need 300kbps of the allotted 320kbps to store all of the required samples.
Now, this isn't exactly how MP3 encoding works, but it is important to note that some recordings may very well be indistinguishable in the 320kbps format, while others will present readily apparent differences.
In my own library, there is plenty of music where I most likely would not be able to differentiate 320kbps MP3 from FLAC. At the same time, there exists music in my library, particularly symphonies, where FLAC is quite clearly superior. Simply put, if you have a symphony recording featuring 80 instruments, there may be too much going on for compression algorithms to correctly extrapolate the full experience.
timestretch
Limp Gawd
- Joined
- Mar 19, 2011
- Messages
- 200
Just curious, TESLA, have you taken the opportunity to ABX some of the symphonic recordings lossless vs lossy?
Foobar2000 and the ABX comparator component make this very easy to do. I think tonight I will try testing some of my symphonic stuff (Vivaldi, Rachmaninoff) in this manner. I hope to pass the test but I have some lingering doubts.
If there is any interest, perhaps we can start a lossy vs lossless, ABX comparison thread.
Foobar2000 and the ABX comparator component make this very easy to do. I think tonight I will try testing some of my symphonic stuff (Vivaldi, Rachmaninoff) in this manner. I hope to pass the test but I have some lingering doubts.
If there is any interest, perhaps we can start a lossy vs lossless, ABX comparison thread.
When I get off of work I can give it a try.
I remember running a comparison for my own curiosity a while back where I would convert random tracks in a symphony to MP3 while keeping the rest as FLAC. I could clearly differentiate the boundaries. (The particular symphonies I used were gapless where one track lead directly into another; the piece was broken up into tracks for convenience only.)
Keep in mind that when I say that I could clearly differentiate, I was doing so in a rather controlled environment. I have a listening chair by a window with a headphone setup on a side table. I was literally just sitting there, sun on my face, with my eyes closed trying to focus on individual instruments, moving back and forth between them.
I would imagine that if I had done the same test while performing a side task, the differences may not have been so apparent.
I remember running a comparison for my own curiosity a while back where I would convert random tracks in a symphony to MP3 while keeping the rest as FLAC. I could clearly differentiate the boundaries. (The particular symphonies I used were gapless where one track lead directly into another; the piece was broken up into tracks for convenience only.)
Keep in mind that when I say that I could clearly differentiate, I was doing so in a rather controlled environment. I have a listening chair by a window with a headphone setup on a side table. I was literally just sitting there, sun on my face, with my eyes closed trying to focus on individual instruments, moving back and forth between them.
I would imagine that if I had done the same test while performing a side task, the differences may not have been so apparent.
I ABX classical music and failed to differentiate. You need to do a large sample for it to be accurate because anyone might guess 50% when you only use 12 samples. MP3 compression works by removing inaudible frequencies. If I can't hear those frequencies I don't need them wasting HDD space. Just think, if I was deaf I could have zero byte MP3 files.
timestretch
Limp Gawd
- Joined
- Mar 19, 2011
- Messages
- 200
A sample of 10 with 8/10 correct would give you a good confidence that you are discerning a difference. Any result near 50% would give a good confidence that you are just guessing. A sample of 8/10 wrong would give a good confidence that you prefer MP3s to flacs AND can tell the difference. Wouldn't that be curious?
And now I am imagining the silent music library of a deaf gentleman audiophile... Blank CDs slipped ritualistically into fine CD players hooked to not-plugged-in loudspeakers.
I don't think 10 is enough because someone may luckily guess 8/10 even. If they did 10 samples every day over a week and they consistently got 8/10 then it would be valid. But I don't see anyone doing that so those that claim they can tell the dif is just hearsay and nothing more.
There is a formula to determine how strong a given probability is based on sample size. I used to know it, but this was some time ago.
The best thing to do would be to get N people together and blindly divide them into two groups. Group one would be the control group and would be asked to listen to M songs, each one twice, all encoded the same way, be it MP3 or FLAC.
So if you had M = 2, the test would consist of:
A1) Song 1 MP3
B1) Song 1 MP3
Which sounds better, A1 or B1?
A2) Song 2 MP3
B2) Song 2 MP3
Which sounds better, A2 or B2?
Then the second group would be your trial group and would listen to the same M songs, but this time the encodings would be mixed:
A1) Song 1 MP3
B1) Song 1 FLAC
Which sounds better, A1 or B1?
A2) Song 2 FLAC
B2) Song 2 MP3
Which sounds better, A2 or B2?
Then you would do some math to find the control group's deviation (using 50% as the expected average) and the trial group's FLAC win vs loss average and deviations and find the probability, using the sample size N and the deviations to determine its strength.
If the probability's strength is significant, we could all accept it and move on, right? Right?
Well probably not, but it would be interesting to see an actual statistical study on this, even if just for kicks.
The best thing to do would be to get N people together and blindly divide them into two groups. Group one would be the control group and would be asked to listen to M songs, each one twice, all encoded the same way, be it MP3 or FLAC.
So if you had M = 2, the test would consist of:
A1) Song 1 MP3
B1) Song 1 MP3
Which sounds better, A1 or B1?
A2) Song 2 MP3
B2) Song 2 MP3
Which sounds better, A2 or B2?
Then the second group would be your trial group and would listen to the same M songs, but this time the encodings would be mixed:
A1) Song 1 MP3
B1) Song 1 FLAC
Which sounds better, A1 or B1?
A2) Song 2 FLAC
B2) Song 2 MP3
Which sounds better, A2 or B2?
Then you would do some math to find the control group's deviation (using 50% as the expected average) and the trial group's FLAC win vs loss average and deviations and find the probability, using the sample size N and the deviations to determine its strength.
If the probability's strength is significant, we could all accept it and move on, right? Right?
Well probably not, but it would be interesting to see an actual statistical study on this, even if just for kicks.
timestretch
Limp Gawd
- Joined
- Mar 19, 2011
- Messages
- 200
I forget the formula as well, but I did take a statistics class back in college.
Anyway, it's extremely unlikely that a person would guess correctly 8/10 times. I think the confidence interval would be something like 90%? That's not good enough for rocket science, though, so let's say 16/20. That's enough questions right on the written driver's license test in my state to qualify for the road test, so that's not too bad.
I guess I can see a person guessing correct 8 out of 10 times, but to maintain that proportion up to 20 trials? That would be the luckiest guesser in town, or a guy who is legitimately noticing a difference of some kind.
I, personally, am not so interested in seeing the ability of a group to discern differences between flac and mp3. People go on and on so much about "golden, practiced audiophile ears" so much that I am content to just test individuals and their own claims, rather than test a bunch of people in a trial group.
Foobar's ABX comparator works like this:
You take 1 lossless song and make a lossy rip of it. Take the 2 tracks and put them into the comparator. For me it was a flac file and a 320kbps mp3 rip of that file.
Then it presents you with trial #1. You have 4 tracks in 2 pairs: A/B, and X/Y. Each pair is one lossy, one lossless. For example, A might be the lossy rip, B is the lossless rip. X is the lossless rip and Y is the lossy rip.You can play any of the 4 for any duration, skip to any point in the track, and switch between them at any moment gaplessly.
Your job is to give 1 of 2 answers. You can say "A is X and B is Y," or you can say, "A is Y and B is X." The way I go about the test is to compare between A and B or a while until I think I know which one is lossless and which one is lossy. Then I do the same for X and Y, match the lossless to the lossless and give my answer. You could also try hearing the difference (or lack of it) between A and X, for example, to arrve at your answer. It seemed easier to listen to things when I knew there was a difference and was trying to find it, though.
Then begins trial 2. Things get randomized and you repeat the same thing. Over and over for as many trials as you like. No time limit per trial.
If you can really tell the difference between lossy and lossless, you'll do well in this test. The idea is that for each trial you need to correctly identify 2 pairs of lossy vs lossless. It's really difficult to just guess.
At least in my opinion, anyone who gets higher than 80% in 10+ trials of the ABX Comparator suggests there's a pretty high likelihood that they hear the difference between lossy and lossless.
I definitely failed my test, though.
Anyway, it's extremely unlikely that a person would guess correctly 8/10 times. I think the confidence interval would be something like 90%? That's not good enough for rocket science, though, so let's say 16/20. That's enough questions right on the written driver's license test in my state to qualify for the road test, so that's not too bad.
I guess I can see a person guessing correct 8 out of 10 times, but to maintain that proportion up to 20 trials? That would be the luckiest guesser in town, or a guy who is legitimately noticing a difference of some kind.
I, personally, am not so interested in seeing the ability of a group to discern differences between flac and mp3. People go on and on so much about "golden, practiced audiophile ears" so much that I am content to just test individuals and their own claims, rather than test a bunch of people in a trial group.
Foobar's ABX comparator works like this:
You take 1 lossless song and make a lossy rip of it. Take the 2 tracks and put them into the comparator. For me it was a flac file and a 320kbps mp3 rip of that file.
Then it presents you with trial #1. You have 4 tracks in 2 pairs: A/B, and X/Y. Each pair is one lossy, one lossless. For example, A might be the lossy rip, B is the lossless rip. X is the lossless rip and Y is the lossy rip.You can play any of the 4 for any duration, skip to any point in the track, and switch between them at any moment gaplessly.
Your job is to give 1 of 2 answers. You can say "A is X and B is Y," or you can say, "A is Y and B is X." The way I go about the test is to compare between A and B or a while until I think I know which one is lossless and which one is lossy. Then I do the same for X and Y, match the lossless to the lossless and give my answer. You could also try hearing the difference (or lack of it) between A and X, for example, to arrve at your answer. It seemed easier to listen to things when I knew there was a difference and was trying to find it, though.
Then begins trial 2. Things get randomized and you repeat the same thing. Over and over for as many trials as you like. No time limit per trial.
If you can really tell the difference between lossy and lossless, you'll do well in this test. The idea is that for each trial you need to correctly identify 2 pairs of lossy vs lossless. It's really difficult to just guess.
At least in my opinion, anyone who gets higher than 80% in 10+ trials of the ABX Comparator suggests there's a pretty high likelihood that they hear the difference between lossy and lossless.
I definitely failed my test, though.
Ah see, you mentioning 10 samples tripped me up as I took that to be 10 individual tracks as the samples with you saying FLAC 8 times.
It's very different if it is 10 * 10 tracks. Then yes, I would agree that it would be rather unlikely for the results to be chance alone.
I still want to ABX the files, but I haven't had the time. Maybe I'll get to them tonight.
Also, looking at the files again now that I am home, they are 24-bit FLAC files. Not CD rips. So it is quite possible that that would make a difference. (Although I still feel like I would have converted them to 16-bit FLAC for my previous comparison.)
It's very different if it is 10 * 10 tracks. Then yes, I would agree that it would be rather unlikely for the results to be chance alone.
I still want to ABX the files, but I haven't had the time. Maybe I'll get to them tonight.
Also, looking at the files again now that I am home, they are 24-bit FLAC files. Not CD rips. So it is quite possible that that would make a difference. (Although I still feel like I would have converted them to 16-bit FLAC for my previous comparison.)
I would not be able to move on because I am forever seeing people saying MP3 sounds crap compared to lossless and they never have done any ABX testing. I have don ABX testing so I can not let BS be posted. I used to say MP3 was crap too until I actually did the ABX testing and proved myself wrong.
Godmachine
[H]F Junkie
- Joined
- Apr 7, 2003
- Messages
- 10,472
Haven't seen such a great discussion on conversion rates of MP3's and the sillyness of some audiophiles and there snake oil claims.
Enjoying it
Personally I can't hear the micro differences (if they exist at all) but what I do hear is blissful , full. Perhaps I don't have the best ears on the planet that can detect a mouse fart in a hurricane from 2 clicks out but I think even a lower end set of cans and a baseline amp/dac can make a world of difference. Everything in between is just more gravy.
Enjoying it
Personally I can't hear the micro differences (if they exist at all) but what I do hear is blissful , full. Perhaps I don't have the best ears on the planet that can detect a mouse fart in a hurricane from 2 clicks out but I think even a lower end set of cans and a baseline amp/dac can make a world of difference. Everything in between is just more gravy.
Unknown-One
[H]F Junkie
- Joined
- Mar 5, 2005
- Messages
- 8,909
Sounds like the first cable had a grounding issue, and the USB device has poor isolation or filtering.
- Data and noisy power are sent out from the PC.
- Data gets passed to the DAC perfectly, and decoded perfectly (Your audio made it over that USB cable to the device with no loss of quality).
- Noisy power coming from your computer runs the pre-amp.
- Perfect digital signal meets noisy pre-amp and gets destroyed.
End result? The digital components don't care much about the cable. The analog components (like the pre-amp) can be effected by the cable.
You might consider using a powered HUB with decent internal power filtering to run USB audio devices, rather than the power coming from your PC.
- Data and noisy power are sent out from the PC.
- Data gets passed to the DAC perfectly, and decoded perfectly (Your audio made it over that USB cable to the device with no loss of quality).
- Noisy power coming from your computer runs the pre-amp.
- Perfect digital signal meets noisy pre-amp and gets destroyed.
End result? The digital components don't care much about the cable. The analog components (like the pre-amp) can be effected by the cable.
You might consider using a powered HUB with decent internal power filtering to run USB audio devices, rather than the power coming from your PC.