Emotiv Hardware
Posted: Fri Dec 30, 2011 6:35 pm
Hello all
This is my first post to this forum...
I have a question regarding the Emotiv headset. I am aware that an openvibe driver has been available for more than a year now but after looking through your posts, I could not find anybody that has reported good or bad experiences.
My interest is in EEG hardware devices and I was given an Emotiv Epoc a year ago for testing/evaluation etc. I have opened and partially back engineered this device and have not been impressed. Specifically, I have found that the input impedance is only 1 megaohm (!) It is so low as to be ridiculous, and I think a few people might not believe me.
For people that are not familiar with EEG amp specifications, a typical amplifier should have at least x1000 higher impedance, i.e. 1Gohm.
If one considers that the Emotiv uses wet saline electrodes (which have rather high and varying impedance) the conclusion is that the said impedances form a voltage divider that is absolutely random and varying. Conclusion: the Emotiv cannot produce reliable voltage amplitude readings. This is disastrous for neurofeedback applications. I don't know much about BCI applications, I guess unreliable signal amplitude is not quite as important here as in NFB, but I would presume BCI apps would perform lousy as well. For example, I don't know the P300 speller algorithms, but I could imagine that the accuracy of such a program would absolutely plummet when using an Epoc for signal acquisition.
My Emotiv is the ordinary $300 consumer version, I cannot say if it's hardware is different to that of the "research" edition, but I doubt it.
Is there anyone having used the Emotiv for more than an hour to "play"? How useful and/or reliable have you found the EEG data to be?
Regards
Stefan
PS. The original Emotiv patents talk about active electrodes. If Emotiv had added the extra opamp to each electrode the above mentioned problem would have be solved 100%. But this was obviously dropped due to cost constrains. As it now stands, I regard the device as unusable, a toy used more to pick up EOG/EMG activity than to monitor (reliable) EEG data.
This is my first post to this forum...
I have a question regarding the Emotiv headset. I am aware that an openvibe driver has been available for more than a year now but after looking through your posts, I could not find anybody that has reported good or bad experiences.
My interest is in EEG hardware devices and I was given an Emotiv Epoc a year ago for testing/evaluation etc. I have opened and partially back engineered this device and have not been impressed. Specifically, I have found that the input impedance is only 1 megaohm (!) It is so low as to be ridiculous, and I think a few people might not believe me.
For people that are not familiar with EEG amp specifications, a typical amplifier should have at least x1000 higher impedance, i.e. 1Gohm.
If one considers that the Emotiv uses wet saline electrodes (which have rather high and varying impedance) the conclusion is that the said impedances form a voltage divider that is absolutely random and varying. Conclusion: the Emotiv cannot produce reliable voltage amplitude readings. This is disastrous for neurofeedback applications. I don't know much about BCI applications, I guess unreliable signal amplitude is not quite as important here as in NFB, but I would presume BCI apps would perform lousy as well. For example, I don't know the P300 speller algorithms, but I could imagine that the accuracy of such a program would absolutely plummet when using an Epoc for signal acquisition.
My Emotiv is the ordinary $300 consumer version, I cannot say if it's hardware is different to that of the "research" edition, but I doubt it.
Is there anyone having used the Emotiv for more than an hour to "play"? How useful and/or reliable have you found the EEG data to be?
Regards
Stefan
PS. The original Emotiv patents talk about active electrodes. If Emotiv had added the extra opamp to each electrode the above mentioned problem would have be solved 100%. But this was obviously dropped due to cost constrains. As it now stands, I regard the device as unusable, a toy used more to pick up EOG/EMG activity than to monitor (reliable) EEG data.