Results 1 to 9 of 9

Thread: Alpine PXA-H800, PXA-H701, and Audison Bit1.1 Testing

  1. #1
    Tester Extraordinaire ErinH's Avatar
    Join Date
    Aug 2010
    Posts
    1,247

    Alpine PXA-H800, PXA-H701, and Audison Bit1.1 Testing

    Testing using the Optical Input:

    Hardware Setup:
    • Test signal was generated with RightMark. Signal was 16 bit, 44.1khz.
    • Beach Audio USB Sound Card used to output signal in Digital Format.
    • Toslink cable used from Beach Audio USB Soundcard to the DSP.
    • Analog out from the DSP to Analog In on the M-Audio 2496 Soundcard.
    • All tests resulted in approximately the same input gain level, which is -1.1dB for both Left and Right channels on the M-Audio 2496 Soundcard. The largest delta is 0.3dB on the H701 test.
      This is key because larger swings in test levels can create less truthful comparisons. The goal was also to make sure I stayed within the DSPs nominal range and my previous testing/experience with these units helped me to know where their max was. All data you see is with DSP's outputs unclipped with a 0dB, 1khz tone.
    • Outputs 3 & 4 for used for L/R outputs on ALL DSPs. All settings were FLAT and defeat was used when available to ensure this.
    • Volume was controlled either by DSP software or DSP Controller.

    Optical Testing Summary:


    Optical Testing Frequency Response:


    Optical Testing Cross Talk:


    Optical Testing Noise:

    Optical Testing Dynamic Range:

    Optical Testing IMD:

    Optical Testing THD:

    Next up, Analog input testing...
    Your ears: The best tools you have... and they're free, too!

  2. #2
    Tester Extraordinaire ErinH's Avatar
    Join Date
    Aug 2010
    Posts
    1,247
    Testing using the Analog Input:

    Hardware Setup:
    • Test signal was generated with RightMark. Signal was 16 bit, 44.1khz.
    • M-Audio 2496 Sound Card used to output signal in Analog Format via RCAs to DSPs.
    • Analog out from the DSP to Analog In on the M-Audio 2496 Soundcard.
    • All tests resulted in approximately the same input gain level, which is -1.1dB for both Left and Right channels on the M-Audio 2496 Soundcard. The largest delta is 0.3dB on the H701 test.
      This is key because larger swings in test levels can create less truthful comparisons. The goal was also to make sure I stayed within the DSPs nominal range and my previous testing/experience with these units helped me to know where their max was. All data you see is with DSP's outputs unclipped with a 0dB, 1khz tone.
    • Outputs 3 & 4 for used for L/R outputs on ALL DSPs. All settings were FLAT and defeat was used when available to ensure this.
    • Volume was controlled either by DSP software or DSP Controller.

    Analog Testing Summary:


    Analog Testing Frequency Response:


    Analog Testing Cross Talk:


    Analog Testing Noise:

    Analog Testing Dynamic Range:

    Analog Testing IMD:

    Analog Testing THD:


    - End
    Your ears: The best tools you have... and they're free, too!

  3. #3
    Tester Extraordinaire ErinH's Avatar
    Join Date
    Aug 2010
    Posts
    1,247
    Thoughts:
    I believe Jim Walter of Alpine said one of their main goals during the design of the H800 was to have a low-noise input stage. Well, it certainly seems they accomplished their goal compared to the 701 and even bitone.
    • The H800 is about 3dB less noisy than the 701 and 6dB less noisy than the bitone. Realistically, though, these values are low already.
    • Intermodulated distortion and THD is lower on the H800 than the other two processors.
    • For analog testing, the FR of the bit1 has a slight bump in the low end while falling between the H800 and H701 in regards to top end FR.
    • For optical testing, the FR is more flat on the H800 than the other two. The bit1 rolls off starting at 6khz and is down 2dB @ 20khz.
    • Optically, the H800 wins out over the other processors again. Lower THD, IMD, Noise, and better FR.
    • You may notice that the Bitone THD & IMD results are odd... really odd. Well, I thought so, too. I re-measured to verify. Same results.
      I went a step further and tested with the input stage set to much lower output. Same result trend.
      I then lowered the output of both L/R channels and lowered the overall volume. Again, same result trend. What you see is what I've verified to be legitimate with my testing. This, again, is why you should use this data for comparative purposes within itself.
      The THD and IMD of the bit1 looks pretty bad. Though, I've personally never had issues with the processor.
      However, if you look only at the summary, the bit1 wins out in THD. All the more reason to use the summary tables AND the graphs together.


    If anyone would like me to also test X processor, I'm happy to do it. You'll have to send it to me and cover shipping back (can't do this stuff on my own dime because it's too cost prohibitive). I've saved all settings and can easily replicate them for future testing; that was one requirement in case someone wanted me to compare another processor later down the line.

    I'd like to thank Jason (ImJustJason) for donating his H701/C701 combo for testing and Frank for his wisdom on the RightMark software.
    I hope you guys enjoy. If you have any questions that I haven't covered here already, feel free to ask. I'll try to answer the best I can.


    - Erin
    Your ears: The best tools you have... and they're free, too!

  4. #4
    Tester Extraordinaire ErinH's Avatar
    Join Date
    Aug 2010
    Posts
    1,247
    Quote Originally Posted by ecbmxer View Post
    How would the AI-net input compare to the analog and digital on either the H701 or H800? Do you think you could do a quick comparo using AI-net between the two?
    Thanks for putting in all of this work!
    Just tested the H800 from the w505 using all three connections:
    1. Optical
    2. Ai-Net
    3. RCA

    Keep in mind these results are given for COMPARATIVE PURPOSE ONLY! Please only look at the data given and compare it within itself. I can't stress this enough.
    The purpose of this is to help see what method of input is best and what the tradeoffs are for different input types. Not to define parameters wholly.

    Testing the H800's Ai-Net, Optical, and RCA Inputs from the Alpine IVA-W505:

    Hardware Setup:
    • Test signal was generated with RightMark. Signal was 16 bit, 44.1khz. Test Signal was burned to CD and played through Alpine IVA-W505 Headunit using the CD source which allows for ai-net and optical output in to the PXA-H800.
    • M-Audio 2496 Sound Card used to output signal in Analog Format via RCAs to DSPs.
    • Analog out from the H800 to Analog In on the M-Audio 2496 Soundcard.
    • All tests resulted in approximately the same input gain level, which is -1.1dB for both Left and Right channels on the M-Audio 2496 Soundcard.
      This is key because larger swings in test levels can create less truthful comparisons. The goal was also to make sure I stayed within the DSPs nominal range and my previous testing/experience with these units helped me to know where their max was. All data you see is with DSP's outputs unclipped with a 0dB, 1khz tone.
    • Outputs 3 & 4 for used for L/R outputs on ALL DSPs. All settings were FLAT and defeat was used when available to ensure this.
    • Volume was controlled either by DSP software.

    Input Comparison Testing Summary:

    Input Comparison Testing Frequency Response:

    Input Comparison Testing Cross Talk:

    Input Comparison Testing Noise:

    Input Comparison Testing Dynamic Range:

    Input Comparison Testing IMD:

    Input Comparison Testing THD:


    - End

    Thoughts:
    From the above, it's seemingly obvious that the optical output wins out. It should be the case that both the ai-net and RCA outputs are influenced by the headunit DAC and, if that is indeed the case, then it's also obvious that the RCA output is the worst way to go out of this deck, assuming you have the ai-net as an option.
    The problem, then, is you're limited to the headunit's DAC.
    Now, some decks may test better. The downside there is you won't have the luxury of the ai-net control. This is a tradeoff. For analog, you may not need the ai-net control; it's really only primarily useful for things such as preset and subwoofer level control for the H800. If that's the case and your deck has a good DAC built in, then you're probably just fine.
    Also keep in mind one downfall of this testing is that the output of the headunit from the RCAs needs to be verified as being unclipped. If I were to toss another random deck in, drive voltage up to meet required test threshold, there's a good possibility that the outputs could be clipping and murdering the results. Luckily, I've owned this deck before and know it's limits because I've tested it (a few times).
    If I were starting with a new deck, I'd test the output at intended test deck volume to make sure it wasn't driven to clipping. Luckily, though, this is really a non-issue; every deck I've tested thus far didn't clip at max volume using a 0db, 1khz track. Can't guarantee they won't all do this, though...
    However, since this is the gear I'll be using in my car for the time being, the obvious choice is to go with an optical output and ai-net output rather than RCA output.
    Your ears: The best tools you have... and they're free, too!

  5. #5
    Tester Extraordinaire ErinH's Avatar
    Join Date
    Aug 2010
    Posts
    1,247
    It's interesting to compare the above results to the p99 results:
    http://www.mobilesoundscience.com/f1...prs-tests-737/

    If I believe the power supply isn't influencing the results the p99 beats out the other products in this comparison thread. Even when using the optical input of these DSP's.
    The only pertinent non-differential I see is the noise levels measured between the H800 and the P99. Additionally, you can see some ipod/cd source comparisons in the above link for the P99.
    If you read this discussion on DAC bypassing linked below, it's interesting to note the alpine decks' source outputs between cd and iPod results differ. In this regard, again, the p99 is the clear winner.

    So, it begs the question: are the limitations of the p99 (single din, no navigation, somewhat limited controls vs alpine's processor/HU combo) going to override the P99's ability to output cleaner signal for both cd and ipod and come in a single package (and in some cases cost much less, depending on which alpine combo you go with)? IMO, it's really up to the user. I do know for a fact that if the P99 was double-din, it would pretty much be everything I'd need in a processor but others may prefer the extra controls given by the other options

    iPod DAC bypass:
    http://www.mobilesoundscience.com/f4...163/#post10939
    Last edited by ErinH; 06-22-2011 at 01:50 PM.
    Your ears: The best tools you have... and they're free, too!

  6. #6
    Junior Member
    Join Date
    Aug 2011
    Posts
    1
    Quote Originally Posted by ErinH View Post
    Testing using the Optical Input:

    Optical Testing Summary:
    Why at processors with use of optical inputs very big IMD???

  7. #7
    Devil's Advocate Adam_MSS's Avatar
    Join Date
    Aug 2010
    Posts
    1,578
    Real Name
    Adam
    That's a good question. When you look at the IMD number for the optical input here it doesn't seem to be in the same range.

    Quote Originally Posted by ErinH View Post
    Input Comparison Testing Summary:
    You don't use science to show that you're right, you use science to become right. - R.Munroe

    The important thing in science is not so much to obtain new facts as to discover new ways of thinking about them. - W.L.Bragg



  8. #8
    MSS Supporting Vendor
    Join Date
    Sep 2011
    Location
    Dayton, Ohio
    Posts
    137
    Real Name
    Aaron Hammett
    It's most probably jitter from the source. Perform a loopback to get a baseline.
    Last edited by envisionelec; 09-19-2011 at 09:54 AM.
    I'm not a complete idiot. Some parts are still missing.

  9. #9
    Senior Member
    Join Date
    Sep 2010
    Location
    Miami
    Posts
    194
    Quote Originally Posted by envisionelec View Post
    It's most probably jitter from the source. Perform a loopback to get a baseline.
    Looking at it again while toggling the graph back and forth, it sure looks like jitter. The increased noise harmonics are symmetrical to the sides of the fundamental instead of to the right like harmonic distortions are.

    No go on the loopback though. Erin's soundcard only has coax i/o's and the usb soundcard he used for testing only has toslink out. (Shoulda got that emu-0404pci or USB Erin :p;))

Bookmarks

Posting Permissions

  • You may not post new threads
  • You may not post replies
  • You may not post attachments
  • You may not edit your posts
  •