Hello, as you already know the problem with slow read/writes for nvme ssd problems.
You said the problem is because veracrypt still supports containers and not only full disk encryption.
I am writing this to make you aware ( in case you aren't) on why you should drop support for containers in favor of fixing the speed problems.
Windows can create virtual hard drives . Right now you could make any virtual hard drive, any size , and encrypt it with bitlocker. This basically pretty much replicates a vera crypt container without the limitation of the actual vera crypt container.
So if anybody still wants to use containers, you can most likely create a virtual disk directly from veracrypt and apply fully disk encryption to it. This way you fix the lack of containers problem and also can fix the low speeds problem.
Of course, plausible deniability still needs to exist so don't get rid of hidden space.
Also we still have no native support of EFI hidden operating system which is clearly possible like others demonstrated, old mrb is almost extinct, everybody uses eufi with gpt nowadays . I feel that these 2 limitations need to be addressed and be on your top list since vera crypt is almost perfect in any other way.
I am willing to donate as soon as i see these 2 things fixed a considerable amount but lately you haven't been much active around . I hope you are well and you can continue doing a great work on the project as is seriously outdated right now . I hope my idea helps you fixing the problems. Thank you !
If you would like to refer to this comment somewhere else in this project, copy and paste the following link:
I disagree with replacing VeraCrypt file container with Windows virtual hard drive since it is not portable for VeraCrypt to use on Linux and Mac OSs and does not include the VeraCrypt hidden volume capabilities.
If you would like to refer to this comment somewhere else in this project, copy and paste the following link:
pretty sure if you create now a virtual disk and encrypt it you can use hidden part too and vera crypt will see it as drive encryption not container .
Linux and mac use different versions of veracrypt driver anyway so it can retain the old options for linux and mac and use the solution i suggested for windows.
Linux and mac usually only work with containers so users won't see as much performance degradation from it as someone using it as system encryption.
It's been over 3 years since the dev knows about this problem and it really needs to be addressed, if you have a better idea share it with us
Last edit: hiddengod 2021-05-18
If you would like to refer to this comment somewhere else in this project, copy and paste the following link:
People use the same file container on multiple OS's as I stated in my original reply. Windows VHD is not portable.
Since the performance slow down for SSD is within the VeraCrypt driver, I do not know if it is possible to modify or rewrite the driver in order to handle the three encryption methods of disk, partition and file container while increasing performance for SSDs.
Perhaps you can try contacting the developers of the defunct CipherShed to see if they are willing to work for free on the VeraCrypt project.
i will try to contact them and hint them on veracrypt. A team with 5+ developers can do wonders for veracrypt just in few months. Mounir either has lost interest either has other problems more important than veracrypt. Is a shame that the best and most complex encryption program is being abandoned (again). Mounir is offline here for over 1 month already
👎
2
👍
1
Last edit: hiddengod 2021-05-19
If you would like to refer to this comment somewhere else in this project, copy and paste the following link:
I think people tend to forget that Mounir as a fulltime job, family obligations and the data center fire that impacted his livelihood and the VeraCrypt project.
I think people should consider how much time it takes to work on VeraCrypt for three OS's including the various Linux variants.
Mounir Idrassi handling the many features of VeraCrypt across three OS platforms (Linux, Mac, Windows) by performing development, testing, bug fixes and releasing software during his free personal time after working a fulltime job and family commitments for little to no monetary compensation. Just to compile and perform a release takes two days for a release of the three OSs.
Beta software is released to the VeraCrypt community with the hope that users are willing to test new upcoming versions to find issues before the next version is released since it is impossible for the developer to have test machines from the various PC makers, various Windows, Linux, Mac OS versions and various disk partitioning configurations for testing.
Hence, the developer relies on the VeraCrypt user community to help test the beta releases.
As with many Open Source projects, the developers rely on the user community to help each other with issues due to the overwhelming requests for assistance.
TL:DR
Each VeraCrypt user should consider how much work effort is involved on an Open Source project supporting three OSs and their various versions for little financial gain by performing development, testing, bug fixes and releasing software during their free time after meeting their obligations of fulltime job and family obligations. Not to mention some VeraCrypt users feeling entitled to have their support issues and/or features worked on immediately. :)
👍
5
If you would like to refer to this comment somewhere else in this project, copy and paste the following link:
Hi all.
Thank you @hiddengod for your proposal. Thank you also @enigma2illusion for sharing your thoughts and your support.
I will first start by saying that I didn't loose interest in VeraCrypt. The past months have been indeed challenging for me for many reasons, including family issues and the OVH fire. There is also the impact of the dramatic changes in the way of life following the Covid-19 measures by government and which affects my ability to allocate free time to the project.
One thing that few people are aware of is that new features in VeraCrypt are the result of weeks and months of experimentation and testing. There is often a lot of research involved where new techniques are learned or created. The absence of commits or new releases doesn't mean that no work is being done. Moreover, in order to make the most of my free time, I regularly disconnect from forums and social platforms in order to focus on the research activities.
Concerning the NVME SSD speed issue, I started many months ago working on a prototypes for a new VeraCrypt driver where the file container logic is separated from disk logic. It is still work in progress with many issues and different possible solutions.
I agree that with the help of few engineers it will be possible to solve these issues and make VeraCrypt much better but during all these years very few people offered technical contributions and none on the driver part which is critical here. I already reached to some people for help on finding contributors but no body has come forward.
I don't blame anyone for not helping. Everyone of us has constraints and obligations, and working for free for an open source project is not always affordable.
Anyway, work on the speed issue is ongoing, albeit slow, and separating container handling from disk handling is the best approach. I can't promise an ETA but I will see if I can publish an alpha that works without crashing too much.
Last note: I wanted to publish a new release two months ago with latest changes, especially ARM64 support and MSI installer on Windows and XML-language support for UI on Linux but life events disturbed by plan. My priority is to get this version out in order to focus on long standing features.
👍
7
If you would like to refer to this comment somewhere else in this project, copy and paste the following link:
thank you for everything, sad to hear other devs don't want to help. Wish i had the necessary skills .
Hopefully by the end of the year we can see a beta version.
If you would like to refer to this comment somewhere else in this project, copy and paste the following link:
May I suggest admin make Mounir's comments, or this thread a sticky, so this recurring question can be addressed until he is ready to release the version. Best wishes to continued good work and success.
Last edit: Robert iXj Smith 2021-05-21
If you would like to refer to this comment somewhere else in this project, copy and paste the following link:
I knew you were trying out new features and bug-testing it on your own computers first before committing the changes, I just knew it. :) But I think sometimes that the small fixes at least show that you around and care, like documentation or explaining the help commands in the rescue disk screen, but then I have also thought about putting up changes on github myself, but then I'd have to research how to run git on my Windows system, how to make a pull request etc and I don't have the Windows assembler on my system either, because that requires registration. It's at least nice to hear from you once a month if you were to give an update on what's happening with you and the struggles you are facing. And personally, if I had a super-important issue with my system, I think it would be very worthwhile to have direct support from you for cash. I mean, why not, right? I think that you still do support for money, right? Perhaps open-source is great in theory, but when it comes down to it, people need cash, need to eat. Thank you for continuing to work on this project which otherwise would just about be abandoned. It's a curiosity as to why there are other developers who are not contributing so much to the project.
If you would like to refer to this comment somewhere else in this project, copy and paste the following link:
Thanks Mounir IDRASSI, I agree with Enigma2Illusion, and I am thankful for your hard work and I understand that your family comes first. I am very thankful veracrypt is cross platform.
If you would like to refer to this comment somewhere else in this project, copy and paste the following link:
Hello, as you already know the problem with slow read/writes for nvme ssd problems.
You said the problem is because veracrypt still supports containers and not only full disk encryption.
I am writing this to make you aware ( in case you aren't) on why you should drop support for containers in favor of fixing the speed problems.
Windows can create virtual hard drives . Right now you could make any virtual hard drive, any size , and encrypt it with bitlocker. This basically pretty much replicates a vera crypt container without the limitation of the actual vera crypt container.
So if anybody still wants to use containers, you can most likely create a virtual disk directly from veracrypt and apply fully disk encryption to it. This way you fix the lack of containers problem and also can fix the low speeds problem.
Of course, plausible deniability still needs to exist so don't get rid of hidden space.
Also we still have no native support of EFI hidden operating system which is clearly possible like others demonstrated, old mrb is almost extinct, everybody uses eufi with gpt nowadays . I feel that these 2 limitations need to be addressed and be on your top list since vera crypt is almost perfect in any other way.
I am willing to donate as soon as i see these 2 things fixed a considerable amount but lately you haven't been much active around . I hope you are well and you can continue doing a great work on the project as is seriously outdated right now . I hope my idea helps you fixing the problems. Thank you !
I disagree with replacing VeraCrypt file container with Windows virtual hard drive since it is not portable for VeraCrypt to use on Linux and Mac OSs and does not include the VeraCrypt hidden volume capabilities.
pretty sure if you create now a virtual disk and encrypt it you can use hidden part too and vera crypt will see it as drive encryption not container .
Linux and mac use different versions of veracrypt driver anyway so it can retain the old options for linux and mac and use the solution i suggested for windows.
Linux and mac usually only work with containers so users won't see as much performance degradation from it as someone using it as system encryption.
It's been over 3 years since the dev knows about this problem and it really needs to be addressed, if you have a better idea share it with us
Last edit: hiddengod 2021-05-18
People use the same file container on multiple OS's as I stated in my original reply. Windows VHD is not portable.
Since the performance slow down for SSD is within the VeraCrypt driver, I do not know if it is possible to modify or rewrite the driver in order to handle the three encryption methods of disk, partition and file container while increasing performance for SSDs.
Perhaps you can try contacting the developers of the defunct CipherShed to see if they are willing to work for free on the VeraCrypt project.
https://sourceforge.net/p/veracrypt/discussion/general/thread/6ecab98a30/#7d1c
i will try to contact them and hint them on veracrypt. A team with 5+ developers can do wonders for veracrypt just in few months. Mounir either has lost interest either has other problems more important than veracrypt. Is a shame that the best and most complex encryption program is being abandoned (again). Mounir is offline here for over 1 month already
Last edit: hiddengod 2021-05-19
I think people tend to forget that Mounir as a fulltime job, family obligations and the data center fire that impacted his livelihood and the VeraCrypt project.
I think people should consider how much time it takes to work on VeraCrypt for three OS's including the various Linux variants.
Mounir Idrassi handling the many features of VeraCrypt across three OS platforms (Linux, Mac, Windows) by performing development, testing, bug fixes and releasing software during his free personal time after working a fulltime job and family commitments for little to no monetary compensation. Just to compile and perform a release takes two days for a release of the three OSs.
Beta software is released to the VeraCrypt community with the hope that users are willing to test new upcoming versions to find issues before the next version is released since it is impossible for the developer to have test machines from the various PC makers, various Windows, Linux, Mac OS versions and various disk partitioning configurations for testing.
Hence, the developer relies on the VeraCrypt user community to help test the beta releases.
As with many Open Source projects, the developers rely on the user community to help each other with issues due to the overwhelming requests for assistance.
TL:DR
Each VeraCrypt user should consider how much work effort is involved on an Open Source project supporting three OSs and their various versions for little financial gain by performing development, testing, bug fixes and releasing software during their free time after meeting their obligations of fulltime job and family obligations. Not to mention some VeraCrypt users feeling entitled to have their support issues and/or features worked on immediately. :)
Hi all.
Thank you @hiddengod for your proposal. Thank you also @enigma2illusion for sharing your thoughts and your support.
I will first start by saying that I didn't loose interest in VeraCrypt. The past months have been indeed challenging for me for many reasons, including family issues and the OVH fire. There is also the impact of the dramatic changes in the way of life following the Covid-19 measures by government and which affects my ability to allocate free time to the project.
One thing that few people are aware of is that new features in VeraCrypt are the result of weeks and months of experimentation and testing. There is often a lot of research involved where new techniques are learned or created. The absence of commits or new releases doesn't mean that no work is being done. Moreover, in order to make the most of my free time, I regularly disconnect from forums and social platforms in order to focus on the research activities.
Concerning the NVME SSD speed issue, I started many months ago working on a prototypes for a new VeraCrypt driver where the file container logic is separated from disk logic. It is still work in progress with many issues and different possible solutions.
I agree that with the help of few engineers it will be possible to solve these issues and make VeraCrypt much better but during all these years very few people offered technical contributions and none on the driver part which is critical here. I already reached to some people for help on finding contributors but no body has come forward.
I don't blame anyone for not helping. Everyone of us has constraints and obligations, and working for free for an open source project is not always affordable.
Anyway, work on the speed issue is ongoing, albeit slow, and separating container handling from disk handling is the best approach. I can't promise an ETA but I will see if I can publish an alpha that works without crashing too much.
Last note: I wanted to publish a new release two months ago with latest changes, especially ARM64 support and MSI installer on Windows and XML-language support for UI on Linux but life events disturbed by plan. My priority is to get this version out in order to focus on long standing features.
thank you for everything, sad to hear other devs don't want to help. Wish i had the necessary skills .
Hopefully by the end of the year we can see a beta version.
May I suggest admin make Mounir's comments, or this thread a sticky, so this recurring question can be addressed until he is ready to release the version. Best wishes to continued good work and success.
Last edit: Robert iXj Smith 2021-05-21
I knew you were trying out new features and bug-testing it on your own computers first before committing the changes, I just knew it. :) But I think sometimes that the small fixes at least show that you around and care, like documentation or explaining the help commands in the rescue disk screen, but then I have also thought about putting up changes on github myself, but then I'd have to research how to run git on my Windows system, how to make a pull request etc and I don't have the Windows assembler on my system either, because that requires registration. It's at least nice to hear from you once a month if you were to give an update on what's happening with you and the struggles you are facing. And personally, if I had a super-important issue with my system, I think it would be very worthwhile to have direct support from you for cash. I mean, why not, right? I think that you still do support for money, right? Perhaps open-source is great in theory, but when it comes down to it, people need cash, need to eat. Thank you for continuing to work on this project which otherwise would just about be abandoned. It's a curiosity as to why there are other developers who are not contributing so much to the project.
Thanks Mounir IDRASSI, I agree with Enigma2Illusion, and I am thankful for your hard work and I understand that your family comes first. I am very thankful veracrypt is cross platform.