Jump to content

Linus was right.

James

Just to add this in, as it wasn't included in the video.

 

AMD's ECC for Ryzen is not POR (plan of record), which means it isn't post validated. A system can very well say it's running ECC, and it'll show in the options that ECC is running, but that doesn't actually tell you if ECC is enabled. The only way to truly see if it's enabled is to force a bit-flip and see if it catches it.

 

I responded to Torvalds' thread on RWT with this info at the time.

 

https://www.realworldtech.com/forum/?threadid=198497&curpostid=198715

 

Secondary, DDR5 has ECC per chip, not per module. It's quite a different and important distinction designed mostly for the memory cell reliability than correcting errors. Most DDR5 for consumers will be non-ECC, and ECC variants of DDR5 still require that 9th chip to enable true SECDED support.

Link to comment
Share on other sites

Link to post
Share on other sites

Using a parrot to censor Linus' finger.

Nice.

If you can read this you're using the wrong theme.  You can change it at the bottom.

#MuricaParrotGang

The meme thread

 

 

 

All of my image memes are made with GIMP.

 

My specs are crap but if you are interested:

Spoiler

 

The meme-making machine - Optiplex 780:

CPU: Intel Core 2 Duo E8400 @ 3.0 GHz

GPU: NVidia Quadro FX 580

RAM: 2 GB

SSD: Non-existent

HDD: 1 TB

OS: Windows 7

 

Laptop: HP 255 G7

CPU: Ryzen 5 3500U

GPU: Radeon Vega 8

RAM: 8 GB

SSD: 500 GB NVMe

OS: Windows 10

 

Link to comment
Share on other sites

Link to post
Share on other sites

After watching this, didn't you think that either referencing old research on the frequency of Bit-Flips or even running your own experiments was worthwhile? 

 

I'd be extremely curious on how often this actually roughly happens in a modern machine. I have no Idea as I expect not all Bit-Flips would result in a crash.

 

Because references go up to 4 Bit-Flips per day

That seems rather significant...

Link to comment
Share on other sites

Link to post
Share on other sites

If it's so important, why not just use mirroring instead of parity?

Link to comment
Share on other sites

Link to post
Share on other sites

4 minutes ago, xnamkcor said:

If it's so important, why not just use mirroring instead of parity?

Because that would increase cost and reduce performance further, while parity achieves the job well enough already?

F@H
Desktop: i9-13900K, ASUS Z790-E, 64GB DDR5-6000 CL36, RTX3080, 2TB MP600 Pro XT, 2TB SX8200Pro, 2x16TB Ironwolf RAID0, Corsair HX1200, Antec Vortex 360 AIO, Thermaltake Versa H25 TG, Samsung 4K curved 49" TV, 23" secondary, Mountain Everest Max

Mobile SFF rig: i9-9900K, Noctua NH-L9i, Asrock Z390 Phantom ITX-AC, 32GB, GTX1070, 2x1TB SX8200Pro RAID0, 2x5TB 2.5" HDD RAID0, Athena 500W Flex (Noctua fan), Custom 4.7l 3D printed case

 

Asus Zenbook UM325UA, Ryzen 7 5700u, 16GB, 1TB, OLED

 

GPD Win 2

Link to comment
Share on other sites

Link to post
Share on other sites

9 minutes ago, TheYang said:

After watching this, didn't you think that either referencing old research on the frequency of Bit-Flips or even running your own experiments was worthwhile? 

 

I'd be extremely curious on how often this actually roughly happens in a modern machine. I have no Idea as I expect not all Bit-Flips would result in a crash.

 

Because references go up to 4 Bit-Flips per day

That seems rather significant...

I get an average 3/4 bit flips a month on my server and the only reason I know is because I run ECC memory on a btrfs raid array with check-summing enabled.

Link to comment
Share on other sites

Link to post
Share on other sites

45 minutes ago, Ashley xD said:

oh if anyone wants to read Torvalds's full statement, here it is: https://www.realworldtech.com/forum/?threadid=198497&curpostid=198647

Registered specifically to thank you for the link. Always amusing (and informative) reading what Linus (the grumpy programmer one) has to say.

Link to comment
Share on other sites

Link to post
Share on other sites

Someone recently mentioned in a comment somewhere, and I'm starting to notice myself now too, that LTT is more and more focussing on "soupa-gaimahs" stuff. Please don't. I for one don't care about gaming (at least not the way some fanatics respond to this info), but by'n'large tolerate it as it's bleeding-edge tech. I'm sure I'm not the only one. Do more non-gaming content, if only to keep your audience diverse.

 

TIA!

 

Your audience 🙂

"You don't need eyes to see, you need vision"

 

(Faithless, 'Reverence' from the 1996 Reverence album)

Link to comment
Share on other sites

Link to post
Share on other sites

1 hour ago, seee the state im in nooow said:

so it's a little more resilient than non-ECC RAM, but not the ECC anyone expects for servers and scientific computing? o_o
how much less should i worry about errors with the non-DDR5 generation compared to non-ECC DDR4? or rather when should i start worrying about errors in general? (outside of those dealing with finance or accurate results)

 

just to describe my headspace at the moment: after watching the video i kind of decided to drop all my curiosities with ECC on DDR4 and relegate that to the DDR5 generation, and then i've learnt (here) that it's kind of not the full ECC I've had in mind

 

writing as a student in 3D media with a personal threadripper platform, still kind of gauging if i should set aside some eventual curiosity money on ECC RAM. haven't gotten any luck from XMP for a long time (which i think isn't recommended if i prefer stability?)

Check the link I posted on how AMD systems potentially say that 'ECC is enabled' but it's just a register check and there's a chance it isn't actually enabled unless you test for it.

Link to comment
Share on other sites

Link to post
Share on other sites

Registered ECC for the win!

 

NOTE: I no longer frequent this site. If you really need help, PM/DM me and my e.mail will alert me. 

Link to comment
Share on other sites

Link to post
Share on other sites

Sorry, this is some next level bullshit, LMG.

 

Thumbnail: Famous image of LT giving the finger to NVidia

First frame of video: Article of LT saying "F-K you NVidia with said image from thumbnail

 

Video content: Nothing to do with NVidia at all.

 

The line between "appeasing the algorithm" and "actually being disingenuous for clicks" is pretty goddamn thick.

🌲🌲🌲

 

 

 

◒ ◒ 

Link to comment
Share on other sites

Link to post
Share on other sites

My biggest takeawak from this thread is WTF is going with that profile pic of James? He looks possessed.

Main Rig:-

Ryzen 7 3800X | Asus ROG Strix X570-F Gaming | 16GB Team Group Dark Pro 3600Mhz | Corsair MP600 1TB PCIe Gen 4 | Sapphire 5700 XT Pulse | Corsair H115i Platinum | WD Black 1TB | WD Green 4TB | EVGA SuperNOVA G3 650W | Asus TUF GT501 | Samsung C27HG70 1440p 144hz HDR FreeSync 2 | Ubuntu 20.04.2 LTS |

 

Server:-

Intel NUC running Server 2019 + Synology DSM218+ with 2 x 4TB Toshiba NAS Ready HDDs (RAID0)

Link to comment
Share on other sites

Link to post
Share on other sites

47 minutes ago, Master Disaster said:

My biggest takeawak from this thread is WTF is going with that profile pic of James? He looks possessed.

nah its just james

✨FNIGE✨

Link to comment
Share on other sites

Link to post
Share on other sites

3 minutes ago, SlimyPython said:

nah its just james

 

Main Rig:-

Ryzen 7 3800X | Asus ROG Strix X570-F Gaming | 16GB Team Group Dark Pro 3600Mhz | Corsair MP600 1TB PCIe Gen 4 | Sapphire 5700 XT Pulse | Corsair H115i Platinum | WD Black 1TB | WD Green 4TB | EVGA SuperNOVA G3 650W | Asus TUF GT501 | Samsung C27HG70 1440p 144hz HDR FreeSync 2 | Ubuntu 20.04.2 LTS |

 

Server:-

Intel NUC running Server 2019 + Synology DSM218+ with 2 x 4TB Toshiba NAS Ready HDDs (RAID0)

Link to comment
Share on other sites

Link to post
Share on other sites

10 hours ago, Arika S said:

Sorry, this is some next level bullshit, LMG.

 

Thumbnail: Famous image of LT giving the finger to NVidia

First frame of video: Article of LT saying "F-K you NVidia with said image from thumbnail

 

Video content: Nothing to do with NVidia at all.

 

The line between "appeasing the algorithm" and "actually being disingenuous for clicks" is pretty goddamn thick.

Because LinusT flipping the finger ain't just about nVidia. Sure, he made it while commenting on nVidia but he is also pissed at Intel for this very topic of the video. I'm sorry you weren't well read enough and jump to conclusion based on a thumbnail and lack of knowledge. 
Check this post by LinusT out. 
https://www.realworldtech.com/forum/?threadid=198497&curpostid=198647

Link to comment
Share on other sites

Link to post
Share on other sites

Am I missing something here?! What does Linus Torvalds getting angry how nvidia refuses to work with him on Linux drivers, and Intel disabling ECC on consumer chipsets have ANYTHING to do with eachother! I mean I get it both are things that Linus Torvalds got angry about, I know nvidia used to make MOBO chipsets, and I know that these companies are not AMD.... But what do these different topics have to do with each other? I clicked on the video expecting a rant along the lines of "NVIDIA HAS DONE IT AGAIN!" rather than "Linus Torvalds Fbombed and flipped the bird to NVIDIA years ago because of drivers. So because of that, let's talk about why Linus T is angry at INTEL because of ECC!"

Edited by Wh0_Am_1
my bd Grammer lol

In search of the future, new tech, and exploring the universe! All under the cover of anonymity!

Link to comment
Share on other sites

Link to post
Share on other sites

7 hours ago, crazzp said:

Because LinusT flipping the finger ain't just about nVidia.

But that image is only ever associated with Nvidia, there is NO way LMG would not have used that image KNOWING this. There has been a lot of hate around NVidia for quite some time now, so it wouldn't surprise me if they used this particular image as "well it's Linus, but people hate NVidia a lot right now, so if we keep it vague people will come to our video thinking it will be more things to beat down on NVidia for"

 

7 hours ago, crazzp said:

Sure, he made it while commenting on nVidia but he is also pissed at Intel for this very topic of the video

So use a different image of him. it's not like the image they used is the one and only image that exists of him

 

7 hours ago, crazzp said:

I'm sorry you weren't well read enough and jump to conclusion based on a thumbnail and lack of knowledge. 

you're right, i'm not "well read" about it, because i think Linus Torvalds is a colossal bag of dicks and I put all of his words and anything he says in the same pile as 98% of twittter, 100% of 4chan. (see: rambling of insane people who i do not care about in the slightest)

🌲🌲🌲

 

 

 

◒ ◒ 

Link to comment
Share on other sites

Link to post
Share on other sites

On 2/8/2021 at 11:24 AM, crazzp said:

Because LinusT flipping the finger ain't just about nVidia. Sure, he made it while commenting on nVidia but he is also pissed at Intel for this very topic of the video. I'm sorry you weren't well read enough and jump to conclusion based on a thumbnail and lack of knowledge. 
Check this post by LinusT out. 
https://www.realworldtech.com/forum/?threadid=198497&curpostid=198647

I actually followed this thread when it came to the news, and I still find the thumbnail and title to be misleading, I clicked in the video thinking it was going to be about nvidia for some reason.

FX6300 @ 4.2GHz | Gigabyte GA-78LMT-USB3 R2 | Hyper 212x | 3x 8GB + 1x 4GB @ 1600MHz | Gigabyte 2060 Super | Corsair CX650M | LG 43UK6520PSA
ASUS X550LN | i5 4210u | 12GB
Lenovo N23 Yoga

Link to comment
Share on other sites

Link to post
Share on other sites

A little bit Meta, but does anybody know the name of the second background music track from the video? I really like it.

Link to comment
Share on other sites

Link to post
Share on other sites

Create an account or sign in to comment

You need to be a member in order to leave a comment

Create an account

Sign up for a new account in our community. It's easy!

Register a new account

Sign in

Already have an account? Sign in here.

Sign In Now

×