3 years to do a .1 upgrade?

[SEC]

I have a Asus ROG Strix that cannot run SpinRite as it will not boot anything other that MSDOS/DOS from a USB Drive. Been waiting since 2019 for a fix from Steve Gibson and finally 6.1 is released. Total disappointment.
Is SpinRite a dead product? No... really, that is an honest question. GRC has known for years now that many newer PC's cannot run SpinRite and after 3 years since the last update, the problem is still has not been addressed and all that has occurred is a 6.0 to 6.1 update. I look at Steve's blog and there has not been a new post since 2021 and you never see him respond to forum posts. The VIrtualBox solution is ridiculous when you consider that we purchase this product to do a specific job, not to spend more time and money to jury rig something up just to try to find a way to make it work or to have to purchase another old PC just to be able to use SpinRite... it's rather ridiculous.
If SpinRite is as dead a product as it appears to be, does someone out there have a recommendation for an alternative tool that will work on new PCs?

 

[DanR]

Asus ROG Strix

This appears to be a Windows 11 gaming PC. As such it is UEFI boot only and cannot boot/run SpinRite 6.1, which requires a BIOS-DOS boot environment.

However, there may be a way to work around that to get SpinRite 6.1 running on that machine. See the first post in this thread:

https://forums.grc.com/threads/how-to-run-spinrite-on-a-uefi-only-machine-part-1-of-5.1613/

It is NOT a trivial exercise and thus likely not for the faint of heart. But it has worked on many UEFI systems.

 

[PHolder]

finally 6.1 is released. Total disappointment.

Give me your address and I will send you a personal cheque for the cost of the upgrade from 6.0 to 6.1

Don't be fooled by the version number only going up by .1 . SpinRite 6.1 is a major rewrite and has works much faster on machines equipped with hardware it has new drivers to support.

I too would dearly like a version that could be used on my newer PCs and would be bootable under UEFI. The issue is that SpinRite wants EXCLUSIVE access to the hardware, and that's very difficult to do on modern OSes that easily boot on modern machines. This is why SpinRite comes with its own OS version (FreeDOS) and unfortunately there aren't a lot of other choices that @Steve can use, from a license perspective. He did purchase a new OS (at great expense) to use with SpinRite 7.x and one hopes that will come along one day, but Steve is a one man shop and he's got a lot of competing priorities.

I get it, it's annoying that Intel has decided newer PCs should no longer support the BIOS. That's not really Steve's fault, and he has a legacy base of users who rely on SpinRite 6.1 still working in the same way as 6.0 did. This is apparently not you. One presumes you would have refunded your purchase if you never got 6.0 to work for you, so I don't see why you're in any position to complain that 6.1 works the same way. What has happened is YOUR needs have moved on, and SpinRite might not meet them any more (for the present anyway.)

 

[cschuber]

To address the Intel comment above, UEFI is not an Intel requirement. The UEFI Consortium sets the standard. Yes, Intel developed the initial EFI spec, but that was to standardize the interface because the BIOS world was the Wild West.

Addressing the rest of the thread, speaking as an Open Source developer, legacy BIOS will disappear. Just like 16-bit operating systems are virtually gone and just as most Linux distros have deprecated and now de-orbited 32-bit support, as has Windows. And other non-Linux Open Source O/Ss are doing the same -- I'm one of the proponents for removal of 32-bit support because maintaining code that supports BOTH 64-bit and 32-bit is a PITA and waste of time. It's called progress. Let's get with it.

I would imagine at some point Steve might have to desupport legacy PCs in his product. A developer can't maintain the kitchen sink. It's a waste of the developer's time.

Development takes time. When you have thousands of developers working on a project, like the Linux ecosystem, yeah, code is developed rapidly. Smaller projects like reactOS, haiku, the various BSDs, and the like have fewer people and thus take longer to support new hardware. A one man operation will take much longer and will require many more resources, like purchasing hardware he won't ever use again, to painfully reverse engineer undocumented interfaces and work out the niggling bugs.

It takes time. And, don't be disappointed when Steve finally decides to move on and desupport legacy BIOS.

BTW, 32-bit UEFI will also disappear. The current UEFI standard is 64-bit. Your older UEFI O/S may not support newer hardware either. It's progress. Get used to it.

 

[PHolder]

UEFI is not an Intel requirement

I never presupposed it was, but they were the mover on the move to end support for BIOS

https://uefi.org/sites/default/files/resources/Brian_Richardson_Intel_Final.pdf

 

[Scott]

……The issue is that SpinRite wants EXCLUSIVE access to the hardware, … This is why SpinRite comes with its own OS version (FreeDOS) and unfortunately there aren't a lot of other choices that @Steve can use, from a license perspective.

I had earlier suggested that Steve could write a UEFI app which would provide exclusive access to the hardware and require no licensing. There are a few other utilities that boot that way like MemTest86. To date that idea has not gotten traction.

 

[cschuber]

I had earlier suggested that Steve could write a UEFI app which would provide exclusive access to the hardware and require no licensing. There are a few other utilities that boot that way like MemTest86. To date that idea has not gotten traction.

Ahh, an ms-dos-like O/S that allows apps complete control of the hardware? If yes, you don't understand the amd64 aka x86_64 architecture. i.e. The segment registers don't point to addresses, AND, in 64-bit mode we don't use segment registers anymore, they point to what is essentially the start of the address space. In other words (repeating), they're not used. This is an impossible ask.

 

[Steve]

Good discussion, guys. I certainly understand @SEC's thread-starter complaint. I'm no more happy with our current state than he is. Hindsight has more to offer me recently than it has for a long time. Such is life. And the old expression of “if you're not making mistakes you're not trying hard enough” would suggest that I've been trying plenty. So... Would I spend 7 years on SQRL if I knew then what I know now? Hell no. I'm not totally nuts. And I'm not happy with the fact that 3.5 years went into SpinRite v6.1 — a product that can only boot over BIOS-based firmware and still uses the BIOS for USB-connected drives. But in the case of v6.1, I was honoring an aging commitment that I had to fulfill before I could do anything else. As it is, I've learned a LOT from watching the roll-out of v6.1 which has changed my plans for the next SpinRite.

At this moment I'm working to re-establish GRC's contact with the past 20 years of our customers. Long ago I saw the power of communication, and it's something I allowed to die. That was a mistake I'm in the process of correcting.

Once email is reestablished I have two projects planned before I return to SpinRite. I want to create an inexpensive "Pro" version of GRC's most popular freeware, the DNS Benchmark. And then I want to create “Beyond Recall” — GRC's take on secure media content deletion. I know it's been done before... but that's what I was told when I first announced my intention to create a hard disk drive utility. I was told "there are already lots of those". Today, mine is the only one that still exists. And I believe that I have something new to offer.

“Beyond Recall” is significant since it will require me to dig down deeper into the Windows I/O system then I ever have. I wrote a low level network packet driver years ago, so I've been down there before... But it's been a while. The point is that I expect that the work on Beyond Recall will pave the way for the creation of a Windows-hosted v7 of SpinRite. A Windows-hosted SpinRite would not be able to operate on the system's boot drive. But I've seen that many many v6.1 owners are wishing to hook drives to their machines via USB. And if SpinRite simply ran under Windows, it would solve the BIOS vs UEFI problem. And then, since I own a Windows 32-bit emulating OS (RTOS-32) that can boot over either BIOS or UEFI, a "SpinRite Pro" version could add hardware bootability to provide full drive coverage.

Anyway... This is all just to say that having emerged from the incredibly expensive 7-year squander of SQRL and the 3.5-year commitment fulfillment of SpinRite v6.1 ... The famous line from Monty Python “It's just a flesh wound” comes to mind ... and I have no plans to slow down.

We have a LOT more R&D fun in store!

 

[jay thompson]

“Beyond Recall”

Don't charge too much or it will cheaper to send drives to Hillary Clinton for destruction.

 

[Andrew Smith]

At this moment I'm working to re-establish GRC's contact with the past 20 years of our customers. Long ago I saw the power of communication, and it's something I allowed to die. That was a mistake I'm in the process of correcting.

And as a previous purchaser of SpinRite v6.0 it was a great pleasure to hear from Steve. GRC is a long standing part of the very essence of what makes the internet so awesome.

 

[cschuber]

Good discussion, guys. I certainly understand @SEC's thread-starter complaint. I'm no more happy with our current state than he is. Hindsight has more to offer me recently than it has for a long time. Such is life. And the old expression of “if you're not making mistakes you're not trying hard enough” would suggest that I've been trying plenty. So... Would I spend 7 years on SQRL if I knew then what I know now? Hell no. I'm not totally nuts. And I'm not happy with the fact that 3.5 years went into SpinRite v6.1 — a product that can only boot over BIOS-based firmware and still uses the BIOS for USB-connected drives. But in the case of v6.1, I was honoring an aging commitment that I had to fulfill before I could do anything else. As it is, I've learned a LOT from watching the roll-out of v6.1 which has changed my plans for the next SpinRite.

At this moment I'm working to re-establish GRC's contact with the past 20 years of our customers. Long ago I saw the power of communication, and it's something I allowed to die. That was a mistake I'm in the process of correcting.

Once email is reestablished I have two projects planned before I return to SpinRite. I want to create an inexpensive "Pro" version of GRC's most popular freeware, the DNS Benchmark. And then I want to create “Beyond Recall” — GRC's take on secure media content deletion. I know it's been done before... but that's what I was told when I first announced my intention to create a hard disk drive utility. I was told "there are already lots of those". Today, mine is the only one that still exists. And I believe that I have something new to offer.

“Beyond Recall” is significant since it will require me to dig down deeper into the Windows I/O system then I ever have. I wrote a low level network packet driver years ago, so I've been down there before... But it's been a while. The point is that I expect that the work on Beyond Recall will pave the way for the creation of a Windows-hosted v7 of SpinRite. A Windows-hosted SpinRite would not be able to operate on the system's boot drive. But I've seen that many many v6.1 owners are wishing to hook drives to their machines via USB. And if SpinRite simply ran under Windows, it would solve the BIOS vs UEFI problem. And then, since I own a Windows 32-bit emulating OS (RTOS-32) that can boot over either BIOS or UEFI, a "SpinRite Pro" version could add hardware bootability to provide full drive coverage.

RTOS-32 will be desupported later this year (https://www.real-time-systems.com/on-time-rtos-32/). There has also been discussion some corners of the deprecation of 32-bit UEFI.

And, the announcement by Intel of X86S (though this may be derailed by their competition refusing to go down that path). If Intel does get their way, I can see all new Intel X86 chips only supporting 64-bit mode (booted by a 64-bit UEFI BIOS). Supporting a single architecture (amd64) instead of four -- 16-bit (8086), 16-bit protected (286), 32-bit (386), and 64-bit architectures -- would certainly be more cost effective for Intel given that the first three architectures supported by today's chips are rarely used these days.

IMO deprecating 16- and 32-bit support, supporting these modes with older versions of Spinrite, and supporting only 64-bit going forward would save you, the developer, a lot of time and expense. Especially since the handwriting is already on the wall regarding legacy architectures and modes. As a developer, I would certainly consider this option myself. Simplicity does reduce the likelihood of all kinds of bugs and reduces time in development.

Anyway... This is all just to say that having emerged from the incredibly expensive 7-year squander of SQRL and the 3.5-year commitment fulfillment of SpinRite v6.1 ... The famous line from Monty Python “It's just a flesh wound” comes to mind ... and I have no plans to slow down.

We have a LOT more R&D fun in store!

BTW, if Spinrite will only run under Windows, what about those of us who don't have any Windows on our computers? I know a bunch of us who don't.

 

[Steve]

FWIW, I'm unconcerned about the loss of 32-bit support. That's not going to happen anytime soon since there's too much legacy code that won't run otherwise. 16-bit support is still present, as is real mode. They might both be disappearing, but I cannot see that happening with 32-bit code.

I'm also unconcerned with the lack of support for RTOS-32. I purchased and own the source code for RTOS-32, and the idea that everything everywhere always needs "support" is a sad commentary on the state of the world. No one will be surfing the web with GRC's version of RTOS. So I expect that we'll have versions of our stuff that runs natively under Windows, which will be a huge boon to most users, and "Pro" editions that are bootable over either BIOS or UEFI for those who want / need the added power of full hardware access and/or non-Windows operation.

 

[AlanD]

BTW, if Spinrite will only run under Windows, what about those of us who don't have any Windows on our computers? I know a bunch of us who don't.

If you go to the Upgrade page and insert your Transaction code, one of the options it gives you is:-

SpinRite v6.1 Release 4 – This SPINRITE.ZIP is a compressed ZIP archive containing SPINRITE.IMG, a bootable drive image that may be copied onto a USB drive by Linux or Mac users or anyone without access to Windows. Note that the enclosed SPINRITE.IMG file may not be copied into the file system – it is the file system. A special command ('dd') or imaging utility must be used. Until SpinRite v6.1's documentation is online please check GRC's BootAble page or GRC's web forums for help with these copy operations.

 

[PHolder]

SpinRite v6.1

I think cschuber was asking about the future version of SpinRite 7 [Non-Pro] which would be a Windows app.

In answer to that question @cschuber, I believe Steve plans a SpinRite 7 Pro version (presumably at a higher price) which will come with its own UEFI booting capable OS (the real time OS that @Steve previously mentioned.)