Jump to content











Photo
* * * * * 1 votes

Rufus (introduction topic)

rufus

  • Please log in to reply
375 replies to this topic

#26 steve6375

steve6375

    Platinum Member

  • Developer
  • 6597 posts
  • Location:UK
  • Interests:computers, programming (masm,vb6,C,vbs), photography,TV,films,guitars
  •  
    United Kingdom

Posted 23 February 2012 - 07:10 PM

By the way, I forgot to mention - I tested Rufus with a 'fake' 16GB stick that actually has only 2GB of memory inside (mapped 4 times). Your 'check device for bad blocks' was set to 2 passes and it passed!

#27 sambul61

sambul61

    Gold Member

  • Advanced user
  • 1568 posts
  •  
    American Samoa

Posted 23 February 2012 - 07:49 PM

Wow, almost 300 bots are now reading this topic, or may be those are all visitors? :dubbio: :D

#28 Akeo

Akeo

    Frequent Member

  • Developer
  • 301 posts
  •  
    Ireland

Posted 23 February 2012 - 08:22 PM

You seems to have a firm set of coordinates established for the tool, which is good. :)

I most definitively have.

The problem is, your current set merely follows the pathway offered by mainstream at the moment, and may need to be adjusted as you learn along the way.

Aha, but I don't see it as a problem. Mainstream over power-user usage is precisely what I want. If I have to choose between the two, I will always pick mainstream - no contest. Why? Because power users are usually able to handle themselves, so don't require to be helped as much.
I hope this won't be taken the wrong way, but I didn't write Rufus with the reboot.pro community in mind. That's because rebooters (non visitors) are only a limited subset of people I want to see use Rufus.

Keep in mind, what MS offers for mainstream at the moment, was greatly improved and modified by some rebooters here long ago, as they looked far ahead of what's mainstream today. Not surprising, some tools are quite popular here. Not to say, MS can't do it - they simply won't due to marketing strategy, which is very different for devs here. :huh:

Exactly, and you're talking about devs, which I see as a minority, and which Microsoft probably does as well. That the tools specifically written with the very specific reboot.pro community needs in mind (devs or power users, who want a lot more than basic usage) are also the most popular here is hardly surprising at all. Actually, if I wanted to write a tool specifically for rebooters, it'd probably be very close to RMP. But that is not my goal. Whether reboot.pro exists or not, I would probably develop the same tool, because it is based on what I estimate will most benefit mainstream and users at large (though to be fair, and as you previously pointed out, reboot.pro can also help connecting with users at large, which is good).

May I suggest you to look at a concept of not extracting original content from any disk, but instead offering a user to boot any ISO or VHD placed on an empty Thumb? And replace any of them with updated versions easily by simply deleting the ISO file and copying the new one... I know, you're not looking into multiboot (if that's what you mean), but may be your tool prospective users are...

Sorry but I highly doubt so. The way I see it is that the vast majority of people Rufus aims at helping will be interested in one ISO and one ISO only, and most likely an official rather than a custom one.
I very much understand that, as a power user with a specific set of needs, you would like a tool that caters for these needs. Everyone does.
However, unless I see requests proving otherwise, I also believe you are in a minority, and will treat it as such. If I can help you here and there I will. But if I feel that satisfying your or other rebooters very specific needs will be detrimental to mainstream, by adding confusing options for instance, I will put my foot down.

One would usually need several ISO's to service a PC or run other tasks like installing OS from ISO to a removable media, which also requires more knowledge and art than simply copying content from one disk to another. :book:

Yes, power users and sysadmins, whose needs are usually way too specific to produce a generic tool that will satisfy them all (unless one is willing to invest a lot of time).
I don't see much of a middle ground here. Either you go with mainstream and strive to keep things simple for the vast majority, but at the risk of disappointing power users, or you go with power users but create a tool that requires most people to have more knowledge than they'd like about a simple task they want to perform.
As I want to go with the majority, because this is where I feel people need to be helped more, I chose the former.

What I would do is trying to boot an unchanged ISO or VHD from a USB Thumb, then trying to install OS from ISO to a bootable USB Thumb.

Which is not something most mainstream users would do. They would install to regular (non removable USB) HDD.

Once you accomplish it as a user (not necessarily by Rufus), you'll better understand how trends on Reboot are different from plain vanilla offered by MS today. Suddenly, your vision may transform one day:

"I view my Rufus as a tool that will allow users to make a bootable USB Thumb and copy a collection of ISO, VHD, VMDK virtual disks to it without any change in a way that would make them bootable from the Thumb, and easily replaceable".

Sorry but that's only catering to power users, and not what I am aiming for. I prefer the UNIX philosophy better:

"Do a simple task (single ISO -> bootable USB) and do it well".

I'm not aware of such tool at the moment, but the need for it is great and will only grow "like snowball" as someone said. :D

I really don't see regular users going around with a bootable USBs that carry multiple OSes or ISOs as a valid future scenario, sorry. As I said, for most people, installing an OS or running a live distro is a one time process only, therefore, unless there is strong evidence that this is not the case, I will continue to treat is as such. I am well aware that this may not be the case for rebooters, but I see you guys as a minority compared to mainstream users (which I hope you can see as a fair point). Once more, given the choice, I'd rather produce software that disappoints a minority of users, but that the majority are happy with, than the opposite.

Now, there is no guarantee that your vision will change once you tried the above, but at least you'll get better grip of what some other devs are doing here. To share some habits, I use only empty USB Thumbs and HDs with a set of original ISO's and VHDs dropped onto them and replaceable any time, each bootable via OS means or Grub4DOS - our common tool. In fact, I run Win7 and 8 mostly from Diff VHDs instead of HDs, and it offers huge advantages in trying new apps, drivers and solutions without risk. I may be wrong, but don't see any future for tools aiming at fully extracting content of virtual disks onto Flash - this is redundant, and virtual disk multiboot IS mainstream.

I don't see it that way. At all. The way I see it, only sysadmins and OS testers need something like that.
When I use multiple OSes (which I do), I install them on non USB HDDs and use grub or LILO for multiboot. Now, if I was running a datacenter, that would be different, but I am not, and very few people are (re)installing/fixing OSes on a regular basis...
In my view, only sysadmins, datacenter technicians, virus testers and a few others need to go around with OSes that they can quickly restore to vanilla. The rest of us very much try to limit the time we spend having to reinstall an OS... ;)

Again, did you notice: I may be wrong, so shoot for it. OR, consider it a 1st step in your journey... :3th:

Well, I understand your point, and I very much appreciate the exchange of ideas and visions, since it's impossible to predict what users at large will do (Microsoft and other large software companies get it wrong all the time, I also got it wrong more times than I'd like to admit, but then again everybody gets it wrong when trying to predict the future), so I hope that the fact that we currently disagree on the future direction of Rufus is no big deal.

This being said, I like trying new stuff, and see if it can benefit others, so I will definitely have a look at what you think may become mainstream usage. If some of it can be added without to much difficulty and in a manner that is transparent for mainstream users (another dropdown entry besides DOS and ISO for instance), I don't mind doing so. But if it becomes too demanding or if I think it'll detract regular users from performing a simple task in a simple and efficient manner, I'll keep my developer hat and stick with the direction that I see as the best.

#29 Akeo

Akeo

    Frequent Member

  • Developer
  • 301 posts
  •  
    Ireland

Posted 23 February 2012 - 08:33 PM

By the way, I forgot to mention - I tested Rufus with a 'fake' 16GB stick that actually has only 2GB of memory inside (mapped 4 times). Your 'check device for bad blocks' was set to 2 passes and it passed!


LOL! :lol:

Well, first of all, thanks for the test!
Technically, if the 2 GB are good, the drive doesn't actually have bad blocks and the test only checks for bad blocks. As long as a the data written to what the device presents as a block is read the same, we can't exactly say that a block is bad.

More seriously, we are reusing the badblocks code from the e2fsprogs, and more specifically the section that writes the same single byte of data over the whole extent of the device and reads in back. Thus, the test we currently perform is unlikely to detect 'fake' memory sticks. Being able to test for 'fakeness' would mean having data that is random enough to not repeat, yet not too random so as not have to create a 16GB cache file to compare it back. One way would be to use a random block and rotate it on regular basis, but if you're unlucky and the rotation ends up with the same block data at the 'fake' limit, you're not going to see it.
I guess I'll probably try to address it by reserving 4 or 8 bytes to write the block address, at a random location in the block (so that if you run the bad blocks check twice, you would likely ensure that all bytes from a block are properly tested).

I'll log an enhancement request for it, and try to add that feature in a future version. Thanks again for the report.

#30 sambul61

sambul61

    Gold Member

  • Advanced user
  • 1568 posts
  •  
    American Samoa

Posted 23 February 2012 - 09:36 PM

I'd rather produce software that disappoints a minority of users, but that the majority are happy with, than the opposite.

You're free to choose your own ways, which is the most important thing, as our vision tends to change as we learn. Right now your tool seems to address needs of a small minority of Linux distro lovers who likes testing various Distros. Some cite overall Linux popularity at 2% - not sure exactly. But it certainly reflects your personal prefs rather then users' majority. :)

And what you call mainstream is simply users mimicking current consumer MS pathway, because no-one offered a better one, or properly introduced to them. There is absolutely no need to extract any disks anywhere, apart from irrelevant marketing restrictions. Download an ISO directly to USB, add it to the list of bootable - that's it. Meanwhile, MS offers different pathway to businesses (not admins) - a virtual one, and these trends inevitably come to consumer just 2-3 years later. Even faster - with Hyper-V already integrated into Win 8 to serve as common platform for both markets. :lol:

You may consider this site as a mean to popularize your tool, but the main its purpose is to give community chance to learn and get a bit smarter. It shows right away: yesterday a guy was convinced he knows everything, yet today he claims to know just about nothing and reviews all his ideology based on a few knowledge bits happily acquired. Some ignore such opportunity, and they fail despite huge aspirations.

For example, RMPrep USB thread was hit almost 400K times during 2 years - is that mainstream? Not counting Steve's own website hits. Lets see, how many hits your tool will get over the next few months, once initial highlighting period is over. Such reality bits are quite handy to check the course. :rolleyes:

#31 nexusneko

nexusneko
  • Members
  • 1 posts
  •  
    Spain

Posted 23 February 2012 - 09:37 PM

Thak you very much!

Un saludo.
  • Nuno Brito likes this

#32 Nuno Brito

Nuno Brito

    Platinum Member

  • Team Reboot
  • 10447 posts
  • Location:boot.wim
  • Interests:I'm just a quiet simple person with a very quiet simple life living one day at a time..
  •  
    European Union

Posted 23 February 2012 - 10:11 PM

Wow, almost 300 bots are now reading this topic, or may be those are all visitors? :dubbio: :D

I see 500 right now. Wouldn't be so quick to call them bots.

Rufus was featured today at the reboot newsletter.

:cheers:

#33 Me_Too

Me_Too
  • Members
  • 3 posts
  •  
    Portugal

Posted 23 February 2012 - 11:33 PM

I'm not a bot. I can read a CAPTCHA :)

Seems like a nice replacement to good'ol HPUSBFW. In fact it seems to be a little more. It might be a nice replacement for LiLi USB Creator (although I'm surprised that it doesn't support Linux Mint...)

Thank you for the heads up Reboot.pro ;)

#34 sgriso

sgriso
  • Members
  • 1 posts
  •  
    Italy

Posted 23 February 2012 - 11:51 PM

yeeeessss!!! Good one people!!!! Amazing! I love it!

#35 Akeo

Akeo

    Frequent Member

  • Developer
  • 301 posts
  •  
    Ireland

Posted 24 February 2012 - 01:18 AM

Right now your tool seems to address needs of a small minority of Linux distro lovers who likes testing various Distros. Some cite overall Linux popularity at 2% - not sure exactly. But it certainly reflects your personal prefs rather then users' majority. :)

You seem to be forgetting Windows users, whose installation media Rufus supports, and the fact that I stated that I want Rufus to generically support the whole gamut of bootable ISOs (as long as they don't have something specific that renders them incompatible with USB).

Ergo, it doesn't matter if one set of ISOs is used by 2% of people, when most bootable ISOs are going to be supported. Moreover, Rufus supports utilities like Hiren's Boot CD, NT Password Recovery, and other ISOs that wouldn't qualify as Linux distros, but that use the same method a Linux for boot. But even if these accounted for only 2%, you are using a false dichotomy, because a 2% OS market share does not translate in a 2% usage of these OSes with Rufus.

Let's consider for a second that Windows has a 90% market share (which is probably less now, on account of OSX's popularity - also market share is not exactly exclusive, so we may have 1% of these 90% that also account for 1% of the Linux 2%, with a total percentage that's more than 100).

Out of this 90% market share, only a small percentage of users is likely to be interested in an ISO -> USB tool, be it for Windows installation media or Linux installation media.
Personally, I very much doubt that more than 20% or even 10% of all Windows users out there will ever need a tool that converts ISO to USB, since it requires having an use for the resulting drive, which single-OS Windows users wouldn't have outside of recovery (but again recovery tools are currently also served by Rufus' isolinux support, so they should be counted in the Linux share).
Now, out of these 20 or 10%, I genuinely believe that a lot more than 2% will be interested in booting isolinux based USB converted ISOs.

If you are interested in USB -> ISO, you already have a bootable ISO in mind, which, given the current breakdown of ISO support in Rufus, can only be Windows installation media or isolinux based (whether actually Linux or not doesn't matter).

Now, if you want to pretend that 98% of Rufus ISO->USB users will do so with Windows installation media only, be my guest. Personally, I anticipate a much larger percentage than that. In fact, if I were to venture a guess, I'd say it'll probably be close to even.

But then again, I want Rufus to be generic when it comes to ISO, so if I can easily add support for one type of specific ISOs, even if I only estimate that 1% of people will use it, I will do so. Why? Because mainstream users will expect any bootable ISO to be supported, since, very much unlike other tools, the UI does not differentiate in any way between ISOs or distros. But if Rufus fails to support than 1%, when it could fairly easily do so, then I am failing mainstream's expectations. Currently I'm not expecting more than a small percentage of Rufus users to care about older PE based ISOs. Yet I am still planning to adding support for these ISOs to Rufus.

And what you call mainstream is simply users mimicking current consumer MS pathway

Sorry, but you seem to have a very depreciative view of mainstream, and you also seem to think that tools being "restrictive" can only be a result of oversight from developers, rather than careful consideration. Please remember that most people aren't power users, and shouldn't force others to become ones. Considerations such as these are, I believe, the approach that most people who design UIs and tools that, are aimed to the general public at large will tend to take. See the ribbon feature in newer versions of Microsoft Word, which, even despite the outcry from power users, is also going to be carried out into the Windows 8 interface. And what about the removal of the Start button there? (NB: I don't like these features either, but if Microsoft appears to have feedback that more people are finding them useful than decrying it, who am I to judge?).

because no-one offered a better one, or properly introduced to them.

Do you have evidence of that? Did you poll a random set of users (including people who wouldn't frequent reboot pro), let them play with different versions of an application, one that exposed as many features as possible and the other that didn't, and asked them which one they preferred afterwards? Otherwise it's my expectations of what users will want, versus your expectations of what users will want, which can only be based on our existing experiences with software development and support.

There is absolutely no need to extract any disks anywhere, apart from irrelevant marketing restrictions.

Why not? Being able to edit files on the fly can be pretty convenient, especially considering that ISO is aimed at read-only. Again, can you produce evidence that one choice is much better than the other. That it would be more convenient for people interested in multiboot, I have no doubt (but then since it's only copying ISO files, I fail to see why a tool needs to be provided when all that's needed is installing syslinux or grub4dos). I can also see speed gains in copying a large file vs small ones. But apart from that...

Download an ISO directly to USB, add it to the list of bootable - that's it

My understanding is that it can be tricky for some images, since one must have some form or optical drive emulation/virtualization behind the scenes... And that's the reason I chose extraction over single ISO file.

You may consider this site as a mean to popularize your tool, but the main its purpose is to give community chance to learn and get a bit smarter.

If I want to give the community a chance to learn and get smarter, I'd rather go over the Rufus code in details, so that people can produce their own tools, instead of relying on the one produced by others (who may or may not suit their needs). This, IMO, is the best way to empower an IT community, and who knows, it may result in someone producing a tool that I would never have anticipated and yet find very useful. Also, as I stated, not everybody wants to become familiar with the gritty details of the boot process, unless you actually write tools that deal with booting or spend your time helping people who have boot issues.

It shows right away: yesterday a guy was convinced he knows everything, yet today he claims to know just about nothing and reviews all his ideology based on a few knowledge bits happily acquired. Some ignore such opportunity, and they fail despite huge aspirations.

Yes. And please don't assume that I can't produce a grey beard, ragged cloak and thorny stick too. It's not because I'm a newcomer here that I'm green and need to be taught lessons. Also, aren't you sure that, by proselytizing as you do, you aren't actually taking the role of the person who was convinced he knew everything...?

For example, RMPrep USB thread was hit almost 400K times during 2 years - is that mainstream?

OK, then let's check this thread in 2 years, as it'll be the only way to actually have some insight as to whether a 400k page view is significant or not in a comparative context (which won't prevent me from congratulating RMPrepUSB on its apparent success). But if we're gonna compare, I'd rather do it with downloads, because I very much prefer when people to use a software rather than talk about it.
For the record (an please don't misconstrue it as bragging, I don't see these numbers as that impressive), even before I posted about Rufus here, I was getting around 200 downloads of Rufus per day, which I'm quite pleased about. The software is only a little more than 2 months old after all and hasn't seen much advertising so far.

Not counting Steve's own website hits. Lets see, how many hits your tool will get over the next few months, once initial highlighting period is over. Such reality bits are quite handy to check the course. :rolleyes:

Indeed. Can you provide an idea of how many downloads RMPrepUSB sees each day, after 2 years of existence, so that I can get an idea of what I should aim at overtaking then? ;)

#36 Akeo

Akeo

    Frequent Member

  • Developer
  • 301 posts
  •  
    Ireland

Posted 24 February 2012 - 01:28 AM

Seems like a nice replacement to good'ol HPUSBFW.

Thanks for the good words.

although I'm surprised that it doesn't support Linux Mint...

Well, I would say it's not exactly Rufus that doesn't support Linux Mint but rather than Linux Mint doesn't seem to have updated its syslinux/isolinux in a while, whereas Rufus uses the latest version.

If the extra steps needed for Linux Mint and Trinity Boot CD are seen as an inconvenience by many, I'll probably add a fix for their obsolete vesamenu.c32 (in fact I already have an Enhancement Request open for that), though my current concern is that vesamenu.c32 is about as big as Rufus, so I'd rather make it downloadable from the internet if needed, rather than embedded. I also need to log a bug with Mint to see if they can update their syslinux in the next version.

#37 renee

renee

    Member

  • Members
  • 46 posts
  •  
    United States

Posted 24 February 2012 - 04:52 AM

Hi!

I'd like to first of all to congratulate you on a fine product and inform you of some special needs I have. I am technical as I was an OS developer for Digital's VMS engineering. Largely that came to an end when I was hit and run as a pedestrian by an illegal alien. My Dal was killed. The car accident must have aged me by twenty years and I was in the hospital for a year.

A run down on the equipment I have. I have a new extreme with two DVDs in it and a Cruzer 32 GB USB. I presume that there is no need to partition the disk further. But I've started Rufus a couple of times and nothing has happened after I run. I assume it's me. Some error messages may help and some documentation would be great. Is output always in the form of a file? If not...how is the file added to the UFD?

Renee

I am particularly ignorant of the USB segment of the market. I am also ignorant of the technical language used here. So give that I've only been here for a couple of weeks I'm a little last on the basics and I'll give you some examples. I've figured that there's only one source disk.

Edited by renee, 24 February 2012 - 05:07 AM.


#38 Uvais

Uvais

    Frequent Member

  • Advanced user
  • 180 posts

Posted 24 February 2012 - 05:13 AM

waiting for Xp's supported version ;)

Posted Image

#39 renee

renee

    Member

  • Members
  • 46 posts
  •  
    United States

Posted 24 February 2012 - 05:16 AM

Well I see why it didn't run a while ago. I was creating a disk image and thedisk was protected because it had a 1kb file on it.

Renee

Edited by renee, 24 February 2012 - 05:42 AM.


#40 sambul61

sambul61

    Gold Member

  • Advanced user
  • 1568 posts
  •  
    American Samoa

Posted 24 February 2012 - 06:30 AM

Akeo,

I like your agile style. After all an author should be expected to have and defend a set of own coordinates. Unfortunately, I happen to believe, ISO times are almost over. Steve Jobs - remember the guy who offered a first Mac with a Floppy - said several years ago: DVD drive is no more.

Posted Image

Current MS effort is all about OS componentization, resulting in smaller packages targeting certain market segments and easily downloadable without much compression. Their software distribution model is leaning heavily towards VHDs now, which is a logical trend: from full OS install in XP - to OS transfer in 7 - to pre-installed OS distribution on VHD, ready to try and requiring only activation. Install OS from USB won't be a long trend either. PC service from network (network boot) may surpass other methods even in consumer segment in a few years. Given unnoticeable speed differences in running OS from HD and VHD, and trends for mechanical HDs to be replaced by faster SSDs, "simple" means VHD now.

Besides, your concept of booting "any" ISO by unpacking is unrealistic, and would be classified as false advertising for a commercial product. Some of your statements show you are not familiar well with Grub4DOS, which in pair with some drivers can boot most known to be useful in service ISOs without extraction, and install OS from ISO. Instead of full extraction the tool would need to auto add appropriate sections to G4D Menu, and possibly extract small subset of files from certain Linux based ISOs.

ISO as (very restrictive and inconvenient) image format is inevitably destined to vanish following DVD ... a dev must envision that when aiming for future tools. Most consumers noticed that by now, despite some desperate devs invented term "burn the ISO" to keep it alive. A year or two ago many argued 32-bit systems are better for consumer - no-one from trend setters cared. They have own vision, based on grows strategy and efficiency, and one will have to eat it whether you want it or not.

Having said that, persist in making your vision a reality - after all its not that bad compare to some other things people do. :rolleyes: It looks sufficiently highlighted by now, at least for me. What would show genuine interest to your project is number of posts made by nicks with at least 20-50 posts count, not 1 or 2, over the next 2 months - the stats accessible to everyone. :)

#41 BW~Merlin

BW~Merlin
  • Members
  • 6 posts

Posted 24 February 2012 - 07:12 AM

Yup. And it still seems like the tool of choice for people who want to flash a BIOS, which I find weird since people also need to provide external DOS files then.

The exact situation I found myself in the other day when I needed to flash the BIOS of a Dell laptop.  I already had the HP USB format tool and had used it in the past but I couldn't find the DOS files in the rush.  Jumped onto http://www.pendrivelinux.com/ as I use their awesome Universal USB Installer to turn WINPE ISO files into bootable USB and saw "Boot DOS from USB" on the side menu which was exactly what I needed, a couple of second later it was downloaded, running (no installer is great) and I had a bootable DOS USB stick read to BIOS flash.My only suggestion is for tab complete and to be able to skip the keyboard selection screen (probably already an option just haven't read the readme file).Great tool keep up the work.

#42 ianst

ianst

    Newbie

  • Members
  • 20 posts
  •  
    Austria

Posted 24 February 2012 - 09:33 AM

Akeo, I haven't posted in this forum for ages, I recovered the password only to write here:

THANK YOU. Please don't listen critics who claim "there are other tools" when in fact I don't know another tool which matches yours.

I appreciate what "reboot" community does, but I joined the forum only because of ImDisk. Your tool is the second that matches my ideals: having a tool as small and simple as possible but with as much functionality as possible and which has source code. I know a lot of rebooters couldn't care less for the source code but I don't even consider the tool "existing" when it's without it. I know it's subjective.

The GPL3 tool which replaces HP's and does even more was such a great news for me.

Thanks, keep up the good work, greetings from the Alps.

Edited by ianst, 24 February 2012 - 10:07 AM.

  • Nuno Brito likes this

#43 Wonko the Sane

Wonko the Sane

    The Finder

  • Advanced user
  • 13639 posts
  • Location:The Outside of the Asylum (gate is closed)
  •  
    Italy

Posted 24 February 2012 - 09:53 AM

Akeo, I haven't posted in this forum for ages, I recovered the password only to write here:

Well, for the record, your post count is right now at #1, so you haven't ever posted before. :dubbio:

The existence of other tools is not at all a critic to the nice work by Akeo :worship: it is a simple fact, each tool may have better or worse "features", being more aimed to a certain scope or to the other, as always beauty is in the eye of the beholder, but IMNSHO the possibility to choose between many, according to each one's own inclination is undoubtedly freedom :smiling9: (or at least "more freedom than before").

:cheers:
Wonko

#44 Akeo

Akeo

    Frequent Member

  • Developer
  • 301 posts
  •  
    Ireland

Posted 24 February 2012 - 10:30 AM

waiting for Xp's supported version ;)

You mean you've been trying to feed a Windows XP installation ISO to Rufus (PE based), right?
Because otherwise, Rufus does run and was tested on Windows XP, so if you feed it the right ISOs, it should created USB drives there.

I am working on supporting more ISOs, such as XP based ones, but this will take some time...

#45 wimb

wimb

    Gold Member

  • Developer
  • 2281 posts
  •  
    Netherlands

Posted 24 February 2012 - 10:54 AM

Rufus USB Format Utility has nice features such as speed and easy to make NTFS formatted USB-sticks. :)
Also it is easy to make bootable MS-DOS USB-sticks or sticks that can boot with Linux PartedMagic. :thumbsup:

However, it has a very dangerous feature, that is also present in HP format Tool.
When USB-harddisk is connected, it does not recognize partitions. :ph34r:
On formatting it will create new partition for the entire disk,
so that partition table and other data partitions are destroyed (not desirable for the mainstream ).
The problem is that it formats the disk instead of the drive.

When used in XP it does not support bootmgr type ISOs like 7 PE (Failed with Write Error).
In Win7 OS everything worked OK for the same ISO to make USB-stick booting with 7 PE

Also it is a pity that Grub4dos bootloader is not offered as option.
That would allow to boot from the ISO file as it is without unpacking.
The advantage would be 1 reliable file which is easier to maintain instead of a bunch of files.
Also in that case Multibooting would be very easy for everybody, like realized in BOOT_USB.exe
http://www.911cd.net...showtopic=24424

:cheers:

#46 renee

renee

    Member

  • Members
  • 46 posts
  •  
    United States

Posted 24 February 2012 - 11:10 AM

Uh-oh. I think I posted to the wrong thread a few minutes ago. OH Well.

To the young gentlman who received the error message, My guess is that your system cannot find the ISO. It's an educated guess....smile.

Thank you BUT I just found out that Cruzers don't seem to like this application so I'll have to get another USB disk :(.

Renee

Edited by renee, 24 February 2012 - 11:53 AM.


#47 alfreire

alfreire
  • Members
  • 6 posts
  •  
    Spain

Posted 24 February 2012 - 11:34 AM

Tnak you for this great tool... :good:
Regards... ;-)

#48 Akeo

Akeo

    Frequent Member

  • Developer
  • 301 posts
  •  
    Ireland

Posted 24 February 2012 - 11:35 AM

Unfortunately, I happen to believe, ISO times are almost over.

Yes, and I would like to believe that too. I'm also hoping that it will happen eventually. The reality, however, is that, apart from Apple, people who produce OSes or OS-like utilities still deliver them as ISOs.
The future is fine, and will be what it will be, but currently, if you want to install/run Arch Linux, Fedora, NT Password Recovery, Mint, Ubuntu, Slackware, Windows, etc, you will be provided with an ISO only.
This is what the MSDN provides, and this is what Linux distros and utility tools provide. Therefore, as long as companies and people who aren't Apple don't switch to providing something else than an ISO, and that can easily be made to work with USB (which seems to be a challenge, since none of them seem to do - in fact Ubuntu advises to use Universal USB Installer for that, which does the same thing as Rufus), there will be a very real demand for ISO -> USB conversion tools, and probably tools that make the task as simple as possible, because for most people, that part is only one small step in a larger process.

Steve Jobs - remember the guy who offered a first Mac with a Floppy - said several years ago: DVD drive is no more.

Yes, I very much remember the guy. Since I started developing when floppies were a luxury (I only had tape then), of course, I'm a little bit familiar with major figures of the computer industry. I used to like Jack Tramiel's philosophy better: "Computers for the masses, not the classes", because it actually ensured that I could get access to a computer and start tinkering...
I also remember that, like everybody else, Steve Jobs also got it wrong quite a few times (remember the Pippin, Newton and Lisa? Arguably good products yet acknowledged as major flops). I'd very much like him to be right with regards to ISOs being obsolete right now, because then I wouldn't feel like I have to fill a gap with a tool like Rufus, but as I stated above, you just have to look around to find that, no matter how much good old Steve wished for it, it is still a long way from having happened yet.

Also, let me ask you this: How do you feel about the one button mouse? Isn't that also trying to make a product as simple for the users as possible?

Current MS effort is all about OS componentization, resulting in smaller packages targeting certain market segments and easily downloadable without much compression. Their software distribution model is leaning heavily towards VHDs now

I'm not seeing that on the MSDN. Trends are nice, and I don't mind evolution, on the contrary. But if you only address the future market rather than the current one, you won't be of much help to anyone. Also, are you sure you aren't talking about sysadmin trends? Can you got to a computer/software store and buy Windows 7 on VHD? Will you even be able to do so when Windows 8 is out?

There's a difference between systems integrators, like Dell, HP, Lenovo, etc, who have very specific needs, and regular users. Please don't forget that there exists quite a large set of people in the latter group.

"simple" means VHD now.

No. Simple means working with what you actually get now. As far as I could see, neither MSDN or software stores or even Linux distros offer VHD as a distribution media.
Again, you're trying to address a problem that very few people are expected to have right now.

Until VHD overtakes ISO as distribution media for all of the tools and distribution I listed as currently compatible with Rufus, diverting too much time on VHD still seems only like trying to address the needs of a very specific minority.

Besides, your concept of booting "any" ISO by unpacking is unrealistic, and would be classified as false advertising for a commercial product.

How so? As I stated, older ISOs that are natively incompatible with USB are hardly expected to still be used by anyone, and, as long as VHD still doesn't seem to have made any dent into the distribution medium of choice, I fully expect ISO producers to be aware that some may try to use it extracted on USB and want it to work this way. This is exactly what Microsoft went for, and also what other distros creators such as Arch Linux, etc. do.

I was actually pleasantly surprised at how well most of the ISOs I tested seemed to work with Rufus and that I didn't have to spend time playing with grub4dos.

Some of your statements show you are not familiar well with Grub4DOS

I'm not, because I found that I had no need to use it in Rufus yet.

Instead of full extraction the tool would need to auto add appropriate sections to G4D Menu, and possibly extract small subset of files from certain Linux based ISOs.

So there is extraction and customization involved? Might as well extract the whole thing and use the original config, as intended by the distro author, which is the approach I took.
I also had a chat with the Arch Linux distro maker, and he seems to consider extraction and boot of the extracted files from USB as the better choice. He actually went some length to make sure his distro worked this way, even as he is also very much interested in multiboot.

ISO as (very restrictive and inconvenient) image format is inevitably destined to vanish following DVD... a dev must envision that when aiming for future tools.

What good is a tool that leaves present-day users stranded. Again, when VHD becomes the distribution media of choice, I don't have an issue adjusting Rufus for it. But what if something else intervenes or users reject VHD as too complex/too expensive? It looks to me like, despite Apple's best efforts to set the trend, ISO is by a large margin still the distribution media of choice. As such, people very much need ISO -> USB conversion tools. Therefore, I might as well try to address this very current need.

A year or two ago many argued 32-bit systems are better for consumer - no-one from trend setters cared. They have own vision, based on grows strategy and efficiency, and one will have to eat it whether you want it or not.

As a vision, I can't help to find this quite arrogant. If there are "trend-setters", what does that make of the rest of people? Sheep?
Are people who don't care enough about computing to want to be a "trend-setter" not supposed to have any ideas of their own of what they would like their experience to be?

What would show genuine interest to your project is number of posts made by nicks with at least 20-50 posts count, not 1 or 2, over the next 2 months - the stats accessible to everyone. :)

Not at all. What shows genuine interest to the projects are always downloads. I could probably dig up long threads from one place or another filled with posts of how buggy some software is, and plenty of visits, and it will have no bearings on the popularity of the actual software.

#49 renee

renee

    Member

  • Members
  • 46 posts
  •  
    United States

Posted 24 February 2012 - 11:46 AM

Well, I just received one of those non-existant error messages telling me that the Cruzer is not bootable....

Does any one have a recommendation for another USB? I appreciate the whey they are fast.

Renee

#50 Akeo

Akeo

    Frequent Member

  • Developer
  • 301 posts
  •  
    Ireland

Posted 24 February 2012 - 11:48 AM

However, it has a very dangerous feature, that is also present in HP format Tool.
When USB-harddisk is connected, it does not recognize partitions. :ph34r:

Well, the expectation is that most USB hard disk enclosures will not present themselves as removable, so wouldn't be listed in Rufus.
Is that not the case? As much as possible, I tried to make sure that Rufus avoided listing HDDs.

On formatting it will create new partition for the entire disk

Yes. And it also presents a big warning that the content will be destroyed, where users can cancel the operation.
I guess I could add some code that tries to detect if the physical drive contains multiple partitions, and informs the user.

so that partition table and other data partitions are destroyed (not desirable for the mainstream ).
The problem is that it formats the disk instead of the drive.


I have a long standing enhancement request to try to preserve partition, but it is low priority for now, because I thought most HDDs would never be listed in Rufus.

When used in XP it does not support bootmgr type ISOs like 7 PE (Failed with Write Error).
In Win7 OS everything worked OK for the same ISO to make USB-stick booting with 7 PE

Thanks for the report. I thought I had tested with a Win7 installation ISO on XP, but I will test again and address the issue if confirmed.
I logged a new issue for that as well.

Also it is a pity that Grub4dos bootloader is not offered as option.
That would allow to boot from the ISO file as it is without unpacking.

I haven't said I never will. Remember that Rufus is quite young and will evolve. ISO support is only a few weeks old.
I'd like to keep the utility small though, so I may create a separate version that embeds grub4dos. But for now, I see it as low priority.





Also tagged with one or more of these keywords: rufus

3 user(s) are reading this topic

0 members, 3 guests, 0 anonymous users