›

iPad Pro (nov 2018) performance

jfox815jfox815
in General

Does anyone here have the latest iPad Pro (nov 2018)?

I’m getting close to maxing out the number of plugins and instruments my iPad can safely handle without hiccups. I have the 2017 iPad Pro right now. From what I’ve read the 2018 iPad Pro has about twice the performance of the 2017 iPad Pro. Wondering how this translates inside BM3.

Thanks!

Comments

  • tk32tk32

    I know that a few of the users here have the 2018 Pro model, but I'm not sure they'll be able to give you a meaningful comparison with the 2017 Pro (which is what I have) - unless perhaps they've owned both.

    What buffer setting are you using inside BM3? If you can handle the (very slight) increase in latency, setting your buffer to 512 or even 1024 will give you some extra room for more plugins.

    I have my 2017 pro set to 512 almost permanently, as I find this gives a great balance between capacity and speed.

  • jfox815jfox815

    I'm actually wanting to go the other way and have less latency. I know that without using my exact setup it wouldn't be a good judge but I was curious what kind of stress people have been able to put on the new pro models.

    @tk32 said:
    I know that a few of the users here have the 2018 Pro model, but I'm not sure they'll be able to give you a meaningful comparison with the 2017 Pro (which is what I have) - unless perhaps they've owned both.

    What buffer setting are you using inside BM3? If you can handle the (very slight) increase in latency, setting your buffer to 512 or even 1024 will give you some extra room for more plugins.

    I have my 2017 pro set to 512 almost permanently, as I find this gives a great balance between capacity and speed.

  • drezdrez

    @jfox815 said:
    Does anyone here have the latest iPad Pro (nov 2018)?

    I’m getting close to maxing out the number of plugins and instruments my iPad can safely handle without hiccups. I have the 2017 iPad Pro right now. From what I’ve read the 2018 iPad Pro has about twice the performance of the 2017 iPad Pro. Wondering how this translates inside BM3.

    Thanks!

    I have a 2017 13” Pro and because it has a headphone jack, you can Pry it out of my cold dead hands. What are you doing to max it out? Can you be a bit more specific on plugins, tracks, etc?

  • mefismemefisme

    You could also try @winconway 's generalized iOS cpu benchmark for BM3. It's available here:

    http://5pin.link/iab/

  • ocelotocelot
    edited August 2019

    Did you say you wanted to lower the latency (drummer?). Don't forget that latency on iOS is usually set in samples, not milliseconds.

    Edit: I was comparing the 2018 Pro with the 2018 iPad 6, not the 2017 Pro!
    The 2018 Pro should be able to run nearly twice as many AUs at 64 and 128 sample buffers. But there are significant tradeoffs, i.e. stuttering when running only a few AUs, no headphone jack or Home button, etc. Try a store with a good return policy first.

    Here's a nearly-raw data dump (note differences between real-world and stress-tests):
    https://forum.audiob.us/discussion/comment/635360#Comment_635360

  • jfox815jfox815

    Yes, I am a drummer. I’m trying to layer some samples on top of a analog drum. Since it’s live there is some forgiveness for latency but I also don’t want them to “flam” (sorry drummer term).

    @ocelot said:
    Did you say you wanted to lower the latency (drummer?). Don't forget that latency on iOS is usually set in samples, not milliseconds.

    The 2018 Pro should be able to run nearly twice as many AUs at 64 and 128 sample buffers. But there are significant tradeoffs, i.e. stuttering when running only a few AUs, no headphone jack or Home button, etc. Try a store with a good return policy first.

    Here's a nearly-raw data dump (note differences between real-world and stress-tests):
    https://forum.audiob.us/discussion/comment/635360#Comment_635360

  • jfox815jfox815

    For this I’m using BM3 to trigger samples live. I’m using a decent amount of plugins. A lot of DDMf stuff. Probably 10ish audio units and some BM3 plugins on top of that.

    @drez said:

    @jfox815 said:
    Does anyone here have the latest iPad Pro (nov 2018)?

    I’m getting close to maxing out the number of plugins and instruments my iPad can safely handle without hiccups. I have the 2017 iPad Pro right now. From what I’ve read the 2018 iPad Pro has about twice the performance of the 2017 iPad Pro. Wondering how this translates inside BM3.

    Thanks!

    I have a 2017 13” Pro and because it has a headphone jack, you can Pry it out of my cold dead hands. What are you doing to max it out? Can you be a bit more specific on plugins, tracks, etc?

  • AudiogusAudiogus

    Definitely not seeing twice as much according to geekbench. Also, apparently multicore does not count (?).

  • ocelotocelot

    Sure, single-core performance in Geekbench is close between both models, but when I had all those iPads, I noticed a big difference while testing at very low buffers - the 2018 could run nearly 2x as many AUs as the 2017...whoops, actually those were the numbers for the 2018 iPad with the A10, not the A10x. It's been a few months since I looked at my test results in the link I posted.

    While the audio thread in audio apps use only a single core, the other parts like the graphics may take advantage of more than 1 core. The 2018 Pro's Metal and multi-core performance is much higher than the other iPads so I suppose that's the reason it can run more AUs - because of it's 2 to 4 extra cores. Opening many 'heavy' GUIs (like Sugar Bytes) was faster on the Pro, if only by a few milliseconds, but it's noticeable in a high-CPU session. Also tweaking the UI was smoother. It's noticeable, but not significant.

    Plus it depends on the app and AUs too. Some are efficient, others doing the same function are CPU-hogs.

    But again, use a store with a good return policy (like Best Buy).

  • AudiogusAudiogus
    edited August 2019

    @ocelot said:
    Sure, single-core performance in Geekbench is close between both models, but when I had all those iPads, I noticed a big difference while testing at very low buffers - the 2018 could run nearly 2x as many AUs as the 2017...whoops, actually those were the numbers for the 2018 iPad with the A10, not the A10x. It's been a few months since I looked at my test results in the link I posted.

    While the audio thread in audio apps use only a single core, the other parts like the graphics may take advantage of more than 1 core. The 2018 Pro's Metal and multi-core performance is much higher than the other iPads so I suppose that's the reason it can run more AUs - because of it's 2 to 4 extra cores. Opening many 'heavy' GUIs (like Sugar Bytes) was faster on the Pro, if only by a few milliseconds, but it's noticeable in a high-CPU session. Also tweaking the UI was smoother. It's noticeable, but not significant.

    Plus it depends on the app and AUs too. Some are efficient, others doing the same function are CPU-hogs.

    But again, use a store with a good return policy (like Best Buy).

    Yah I have a 2017 10.5 Pro. Home button, headphone jack in the top corner (the position matters most for me, a dongle would be OK I guess if the bloody port was not in the middle), 4 gigs ram, 256 storage, great price used. Anyway, yes it is miles ahead of my 2017 base ipad.

  • ocelotocelot

    Yeah, now I remember thanking you for your advice on the 2017 Pro. Thanks again! Hope you're enjoying elsa.
    And I never should have returned it for the 2018 Pro.

  • AudiogusAudiogus

    @ocelot said:
    Yeah, now I remember thanking you for your advice on the 2017 Pro. Thanks again! Hope you're enjoying elsa.
    And I never should have returned it for the 2018 Pro.

    No prob! We never could do the elsa dance though as we were in different regions.

  • ocelotocelot
    edited August 2019

    edit. I'll make another topic for the glitch kits. Don't belong in this one.

Sign In or Register to comment.