neuromore Studio, new free visual programming language VPL for OpenBCI

wjcroftwjcroft Mount Shasta, CA
edited August 2015 in Software
I exchanged some emails with these folks last November, but just saw today that they have released their version 0.9 with OpenBCI support. I'm unsure yet what their business model is, but they are giving away their Studio package at the moment. Perhaps they will have some sort of cloud subscriptions, Freemium setup, or will license for developers to incorporate their engine, etc. I think one of their goals was tight coupling with external virtual reality / gaming apps. Their team roster looks impressive.

Runs on both Windows and Mac, but only has support for OpenBCI on the Windows version currently. Mac support will come in the next release, probably within a month.

http://www.neuromore.com/blog



A whole slew of demos on that same Youtube channel, cool features.

And... at the same time I'd want to mention Jarek's BioEra software, which has been around for a huge number of years and is the basis for numerous commercial neurofeedback packages; such as LENS, Cygnet, ...  As far as VPL element blocks with exceptional functionality, versatility, BioEra has quite a lead in this field of neurofeedback VPL.

Comments

  • wjcroftwjcroft Mount Shasta, CA
    @Rceballos98 (Rodrigo Ceballos Lentini) posted yesterday a note over on the Community page,

    http://openbci.com/community/neuromore-a-rising-software-acquisition-tool/

    Rodrigo, were you running neuromore on native Windows, or under VirtualBox on Mac?

    William
  • Rceballos98Rceballos98
    Hi William,

    I'm running Bootcamp Win 7 on my Macbook Pro 2015. (Where it runs very smoothly)

    I did try Neuromore on my Win 7 VM but trying to run the Studio leaves a black screen and ramps up RAM usage like crazy. 

    Neuromore devs told me they think the problem stems from VRAM reassignment to different contexts when layouts and widgets are loaded. Apparently, some graphic software doesn't behave well with this reassignment. My old Lenovo Y570 with GT 555M also has the same bug. 

    Neuromore knows about the issue and will keep working on a fix, is my understanding. They have been pretty quick about everything else I've told them so I wouldn't be surprised if this is quickly fixed. 

    OpenBCI support for OS X should also be coming soon. (Next 2-3 weeks)

    Bests,
    Rodrigo
  • nomnomnomnom
    Hi all!

    neuromore Studio dev here. Our Studio doesn't play well with certain GPU/Driver configurations. Most notably the intel 4000 and some related series. NVIDIA and ATI usually works perfectly (if not, please message me!). We mainly develop for Windows 8.x and don't test on Windows 7, Vista or virtual machines. 

    Almost all platform-related problems we encounter are not bugs in our codebase but are caused by QT's hilarious OpenGL quirks. Correction: It's has been going on for such a long time it actually has stopped being funny. Right now it's just sad, but we hope everything goes away with their next QT release(s), as they are in the process of wrapping up some major OpenGL changes.

    If you have any questions or want to give feedback, just message me :)

    - manu
  • wjcroftwjcroft Mount Shasta, CA
    edited August 2015
    Manu, hi

    Isn't it possible that there is SOME Mac virtual machine combo that will run ok on Mac? For example, Rodrigo tried Win 7 with VirtualBox. And had issues with VRAM. But as you say, perhaps Win 8.1 with VirtualBox would have been a better choice. Or Win 8.1 with another VM such as Parallels, VMware, Xen, QEMU, etc.

    Parallels apparently supports user adjustable VRAM settings, this review noticed much better
    performance when the VRAM was set to 1 GB, vs. the default 256 MB. The default is even lower in VirtualBox, 128 MB; and takes some tricks to max it out to 256 MB. That's as far as VBox can go.

    http://blog.parallels.com/blog/2015/03/31/virtual-machine-performance-myths-debunked/

    William
  • nomnomnomnom
    Hi wjcroft!

    We just got the OpenBCI running on OSX today for the first time ever and we will ship OpenBCI support for OSX with the v1.0 release next week :) 

    We aim to support all devices on all platforms so we don't test it on cross-platform VMs. There are very few devices that don't come with an SDK for Windows or OSX, and even if, we can always implement the binary protocol ourselves.

    As soon as we will find the time we will port the studio to Linux, too. Our system is layed out to be ported everywhere and it will happen ;)

  • SirLouenSirLouen
    This is totally awesome, thanks for sharing
  • wjcroftwjcroft Mount Shasta, CA
    v1.0 released August 26 with Mac OS X support for OpenBCI,

    http://www.neuromore.com/blog/2015/8/25/the-next-generation-of-neuroscience-software-neuromore-studio-v10

    Thanks Benjamin, @nomnom (Manu) and your team(!)

  • wjcroftwjcroft Mount Shasta, CA
    edited November 2015
    Benjamin announced v1.1 on Friday Nov 6,

    http://www.neuromore.com/blog/2015/11/6/neuromore-studio-v11-released

    ---

    Bluetooth LE heart rate sensors support for OSX
    We now support devices like the Polar H7 on OSX, so you can use user heartrate in your calculations and also perform HRV analysis on RR intervals.

    New cloud input and output parameter nodes
    Our classifiers can now store single values in the backend and load them again at a later time, so now becomes possible to create things like session counters, remember baselines for following sessions or a parameter that is shared between classifiers.

    New CSV file writer and reader nodes
    Load and save multi-channel data streams in CSV format to your computer, directly from within the classifier.

    New frequency bin selector node
    A new node that allows you to access the individual bins of a frequency spectrum or convert the spectrum into a multi-channel.

    Improved drift correction
    A more flexible drift correction with more settings.
      
    Basic heart rate variability node
    The new HRV Analysis can calculate metrics like RMSSD or pRR50 for bluetooth heartrate sensors like the Polar H7.

    Added Muse Blink and Jaw Clench sensors
    Detect eye blinks and jaw clench events with the muse.

    Added quantile methods to statistics node
    Statistics now support median, quantile and percentile statistic methods.

    Display node errors in classifier editor
    Show the node error directly in the classifier instead of just in the tooltip.

    Settings to enable/disable device detection
    Activate/deactivate individual devices drivers in case you don't need them or they interfere.

    Added Drift correction and more OSC network settings
    We added more parameters to the Studio settings so you can fine tune the drift correction and OSC behavior.

    Redesigned neuromore Cloud
    A heap of invisible changes were made to our backend to make it even more awesome and ready for the future.

  • wjcroftwjcroft Mount Shasta, CA
    Neuromore now has a discussion forum at their site for user questions and news items,


    You need to create a separate username/password for this, rather than using your existing neuromore login. Since the forum is hosted on a separate Wordpress site.

    William
  • spektrolytespektrolyte
    Very excited about the possibilities for Neuromore with it's networked OSC data output, a control framework we have been using in audiovisual projects for a long time now.

    OpenBCI > Neuromore > OSC > LAN > OSC > Resolute Avenue (VJ software) & Ableton Live (audio software) - this looks to be an incredible combination.

    I am just getting started experimenting with the classifiers and need something for data acquisition but don't know much about consumer headsets. Is there anything cheap which will provide enough data to do alpha-theta and SMR feedback testing?


  • wjcroftwjcroft Mount Shasta, CA
    edited December 2015
    Many of the consumer headsets do not provide access to the sites normally used: sensorimotor strip (C3 Cz or C4) for SMR; occipital or parietal sites (O1 O2 or Pz) for alpha or alpha-theta training.

    You could probably rig up your own velcro or elastic band 'headset', then route to a lowcost EEG amp such as OpenBCI, OpenEEG, Pocket-Neurobics, etc. neuromore has support for OpenBCI at the moment.

    Versus has sensors at C3 Cz C4 Fz Pz; so that might be an option for you. neuromore supports one version of that (research Versus?), not sure about the consumer Versus.

    ----

    Emotiv EPOC: AF3, F7, F3, FC5, T7, P7, O1, O2, P8, T8, FC6, F4, F8, AF4
    Emotiv Insight: AF3, AF4, T7, T8, Pz

  • spektrolytespektrolyte
    edited December 2015
    Thanks William for going to the trouble of looking up those sites for me.

    OpenBCI and a set of electrodes looks to be the best bet, but I am struggling to pull the trigger on a new 8ch/32bit with the Ganglion so close and 1/5 the price. 8 channels is a luxury, 4 is enough for neurofeedback with HRV & GSR.

    It's an expensive experiment otherwise, was hoping to prove the principal suggested above without having to invest significantly in hardware. EU import duty on a $600 order is going to be a killer and the wireless option is just an extra unwanted complication.


  • wjcroftwjcroft Mount Shasta, CA
    edited February 2016
    Neuromore is asking for beta testers for their mobile platforms (iOS and Android). You can sign up for access at the link below.

    http://www.neuromore.com/2016/02/16/beta-testing-neuromore-app/

    And a previous blog post with video,

    http://www.neuromore.com/2016/01/22/user-testing-neuromore-app/

    ----

    Also found this December forum post from Benjamin at Neuromore. He states that some time in 2016, there will be two versions of neuromore Studio. The current Community edition which is free. And a Professional edition which has more features.

    Is there a way to play sounds, either as the result of a input (boolean or % of volume), or tied to points? Both would be very helpful!  [bobsmith]

    Unfortunately there is no way to do this within the Community Edition. We already implemented audio triggering based on classifier values as part of our behavior definition system. The feature will be part of the Professional Edition of neuromore Studio and is going to be released next year.  [Benjamin]

Sign In or Register to comment.

© Vanilla Keystone Theme 2024