topbanner_forum
  *

avatar image

Welcome, Guest. Please login or register.
Did you miss your activation email?

Login with username, password and session length
  • Thursday March 28, 2024, 9:53 am
  • Proudly celebrating 15+ years online.
  • Donate now to become a lifetime supporting member of the site and get a non-expiring license key for all of our programs.
  • donate

Last post Author Topic: In search of ... RAMdisk opinions  (Read 33991 times)

IainB

  • Supporting Member
  • Joined in 2008
  • **
  • Posts: 7,540
  • @Slartibartfarst
    • View Profile
    • Read more about this member.
    • Donate to Member
Re: In search of ... RAMdisk opinions
« Reply #50 on: October 01, 2012, 07:56 PM »
Yep, as I mentioned: " Unfortunately, for some reason the WEI score does seem to limit the score to 5.9 unless you have >4GB of memory (no matter how fast the RAM is) - which is plain misleading, IMHO."
OIC. So that  rule is: Must be >4Gb applies regardless of how many channels/RAMcards you have.
So the "exception if you go from single to two sticks because of dual-channel memory architecture" only affects actual RAM speed, not the WEI?
Do I have that right?
It does seem odd.

Stoic Joker

  • Honorary Member
  • Joined in 2008
  • **
  • Posts: 6,646
    • View Profile
    • Donate to Member
Re: In search of ... RAMdisk opinions
« Reply #51 on: October 02, 2012, 06:54 AM »
Perhaps it's a reaction (read CYA maneuver) to the historical tendency of manufacturers to be skimpy with the RAM. Kind of a (pass/fail) visual indicator for the can you do it, yes...should you do it, no hardware choices question.

A bunch of cheap RAM 3-5GB works ok, but trying to skimp with 1 or 2GB of top shelf RAM does not tend to fair well.

f0dder

  • Charter Honorary Member
  • Joined in 2005
  • ***
  • Posts: 9,153
  • [Well, THAT escalated quickly!]
    • View Profile
    • f0dder's place
    • Read more about this member.
    • Donate to Member
Re: In search of ... RAMdisk opinions
« Reply #52 on: October 02, 2012, 07:03 AM »
Some may want to test the weird interesting Bond http://bonddisc.com/ - it is something different from the others; a ram disk is merely a part of it.
I think 'weird' is a good adjective here, Curt - from their documentation:
In contrast to the most common encryption methods used today, which encrypt relatively small blocks of data separately, Bond Disc applies a strong cryptographic algorithm over the entire archive. In contrast todays block-based 128-bit encryption algorithms that encrypt a single block of data in a tiny fraction of a second are increasingly vulnerable to honed brute force attacks from attackers that have massive computing power available. Once a single block is decoded, all data that uses the same key loses its protection.

Bond Disc employs a computing intensive cryptographic algorithm that uses a combination of advanced data scrambling, bit manipulation, far data swaps, variable encryption paths and a strong correlation between the content and key to create an extremely strong encoding of an archive. A change to just a single bit of the content affects the entire encrypted content, that is every byte of it is changed.

That just screams "we're arrogant enough to have designed our own custom encryption scheme instead of using a peer-reviewed standard method" - and that's something that should set off a lot of alarm bells. Their intentions might be good, but there's very++ few people in the world that are qualified to write their own encryption algorithms.

Might still be a decent ramdisk, though? :)

Yep, as I mentioned: " Unfortunately, for some reason the WEI score does seem to limit the score to 5.9 unless you have >4GB of memory (no matter how fast the RAM is) - which is plain misleading, IMHO."
OIC. So that  rule is: Must be >4Gb applies regardless of how many channels/RAMcards you have.
So the "exception if you go from single to two sticks because of dual-channel memory architecture" only affects actual RAM speed, not the WEI?
Do I have that right?
It does seem odd.
Yup, you got it :)

I guess the idea is that one can use solely the WEI numbers to judge a computers performance... but IMHO the mem ops/sec figure becomes useless because they 'corrupt' it with that limitation. And since the installed memory size is a metric that isn't exactly hard to get, it seems even more silly to do this.

Btw, I've been playing around a bit with ReadyBoost on my work laptop, which has a slow harddrive and too little RAM. It does seem to provide some benefit, even though I'm not exactly sure how much is real and how much is placebo. Don't think I see much benefit in things like application launch, but that's probably because the large applications on that system are launched infrequently and sit there for the most of the day, and the small applications are too small to matter... and the whole data working set size is too big to fit in either ram or ram+rbcache. But things like swapping to a program I haven't been using for a couple of hours or resuming from standby (which both seem to incur a fair amount of pagefile access) do seem to leave the laptop groggy for a somewhat shorter while than without.

I've also realized that I have to replace my 2gig corsair flashdrive and probably the 8gig as well - they had been acting up now and then, but with the constant access from ReadyBoost, I've had the 2gig one throw Windows' "the device has malfunctioned" about four times before giving up on it... reinforcing my "usb flashdrives aren't to be trusted!" general opinion, and making me want to warn against putting the pagefile on one (ReadyBoost should be able to handle a drive going tits-up since it's designed as a cache, but I'm pretty sure Windows would BSOD on you if a pagefile went corrupt or missing).
- carpe noctem

f0dder

  • Charter Honorary Member
  • Joined in 2005
  • ***
  • Posts: 9,153
  • [Well, THAT escalated quickly!]
    • View Profile
    • f0dder's place
    • Read more about this member.
    • Donate to Member
Re: In search of ... RAMdisk opinions
« Reply #53 on: October 02, 2012, 07:05 AM »
Perhaps it's a reaction (read CYA maneuver) to the historical tendency of manufacturers to be skimpy with the RAM. Kind of a (pass/fail) visual indicator for the can you do it, yes...should you do it, no hardware choices question.

A bunch of cheap RAM 3-5GB works ok, but trying to skimp with 1 or 2GB of top shelf RAM does not tend to fair well.
Fair enough, but it should have been a different metric. Also note that a score of 5.9 is quite decent, and you could probably get that with a single gig (or even less) of fast ram - the 5.9 cutoff point is "less than 4gig", haven't heard of other cutoff points based on installed ram amount :)
- carpe noctem

40hz

  • Supporting Member
  • Joined in 2007
  • **
  • Posts: 11,857
    • View Profile
    • Donate to Member
Re: In search of ... RAMdisk opinions
« Reply #54 on: October 02, 2012, 07:21 AM »
think 'weird' is a good adjective here, Curt - from their documentation:
In contrast to the most common encryption methods used today, which encrypt relatively small blocks of data separately, Bond Disc applies a strong cryptographic algorithm over the entire archive. In contrast todays block-based 128-bit encryption algorithms that encrypt a single block of data in a tiny fraction of a second are increasingly vulnerable to honed brute force attacks from attackers that have massive computing power available. Once a single block is decoded, all data that uses the same key loses its protection.

Bond Disc employs a computing intensive cryptographic algorithm that uses a combination of advanced data scrambling, bit manipulation, far data swaps, variable encryption paths and a strong correlation between the content and key to create an extremely strong encoding of an archive. A change to just a single bit of the content affects the entire encrypted content, that is every byte of it is changed.

That just screams "we're arrogant enough to have designed our own custom encryption scheme instead of using a peer-reviewed standard method" - and that's something that should set off a lot of alarm bells. Their intentions might be good, but there's very++ few people in the world that are qualified to write their own encryption algorithms.

That's the fly in the ointment IMO.

It's not just a technical or attitudinal issue either.

When implementing security, you always need to ask yourself: Quis custodiet ipsos custodes? Because when it comes to encryption, truer words were never spoken.

You are only as safe as the integrity of the software and its developer.

"Closed source" and "proprietary" are dangerous words when it comes to encryption tools or password management software. It's alarm bells time as f0dder pointed out. :tellme:

f0dder

  • Charter Honorary Member
  • Joined in 2005
  • ***
  • Posts: 9,153
  • [Well, THAT escalated quickly!]
    • View Profile
    • f0dder's place
    • Read more about this member.
    • Donate to Member
Re: In search of ... RAMdisk opinions
« Reply #55 on: October 02, 2012, 07:24 AM »
"Closed source" and "proprietary" are dangerous words when it comes to encryption tools or password management software. It's alarm bells time as f0dder pointed out. :tellme
...which reminds me that I should probably take a week of vacation from work, blow off the cobwebs from fSekrit, shine up the source code a bit, and get it open-sourced.  :redface:
- carpe noctem

40hz

  • Supporting Member
  • Joined in 2007
  • **
  • Posts: 11,857
    • View Profile
    • Donate to Member
Re: In search of ... RAMdisk opinions
« Reply #56 on: October 02, 2012, 07:35 AM »
^ That's ok. We trust you;D

f0dder

  • Charter Honorary Member
  • Joined in 2005
  • ***
  • Posts: 9,153
  • [Well, THAT escalated quickly!]
    • View Profile
    • f0dder's place
    • Read more about this member.
    • Donate to Member
Re: In search of ... RAMdisk opinions
« Reply #57 on: October 02, 2012, 08:27 AM »
^ That's ok. We trust you;D
Should you, though?

*cue x-files theme* :P
- carpe noctem

40hz

  • Supporting Member
  • Joined in 2007
  • **
  • Posts: 11,857
    • View Profile
    • Donate to Member
Re: In search of ... RAMdisk opinions
« Reply #58 on: October 02, 2012, 08:48 AM »
limits_obit.jpg

^Yes we can. Because you're being constantly monitored. By us. On O.B.I.T.  ;D

f0dder

  • Charter Honorary Member
  • Joined in 2005
  • ***
  • Posts: 9,153
  • [Well, THAT escalated quickly!]
    • View Profile
    • f0dder's place
    • Read more about this member.
    • Donate to Member
Re: In search of ... RAMdisk opinions
« Reply #59 on: October 22, 2012, 03:42 PM »
Curt (and the rest of you), Raymond's benchmark has been updated with the most recent version of SoftPerfect (which sports 20-900% speed improvements) - and also, I've added my own quick benchmark.

Verdict: go for SoftPerfect RAMdisk :)
- carpe noctem