Register a SA Forums Account here!
JOINING THE SA FORUMS WILL REMOVE THIS BIG AD, THE ANNOYING UNDERLINED ADS, AND STUPID INTERSTITIAL ADS!!!

You can: log in, read the tech support FAQ, or request your lost password. This dumb message (and those ads) will appear on every screen until you register! Get rid of this crap by registering your own SA Forums Account and joining roughly 150,000 Goons, for the one-time price of $9.95! We charge money because it costs us money per month for bills, and since we don't believe in showing ads to our users, we try to make the money back through forum registrations.
 
  • Post
  • Reply
TheFluff
Dec 13, 2006

FRIENDS, LISTEN TO ME
I AM A SEAGULL
OF WEALTH AND TASTE

gradenko_2000 posted:

https://www.youtube.com/watch?v=qO0-47to8-E

This actually works shockingly well, and you don't even need a pump or have to worry about a pump failing

https://www.youtube.com/watch?v=fMFBAklURCo&t=522s

Swimming pool cooling, while on vacation at the Maldives resort from Hitman 2, because of course. (English subtitles available)

Adbot
ADBOT LOVES YOU

Klyith
Aug 3, 2007

GBS Pledge Week

The Grumbles posted:

I'm aware of how much stuff cost back in the day - I remember 10 years ago building a really good PC for £400-£500 or so - but I guess I'm learning that my idea of what constitutes a rich kid is vastly different from a lot of people in this thread. It was cheaper back then but still felt like a lot of money back then, I could only really afford it because I was working full time but was still a teenager and hadn't moved out of home yet. Maybe it's a country thing?

Sure, but in that case it's not building a PC that's for rich people. It's computers in general. But even back in the 80s you had the IBM PCs that cost as much as a new car, and kit computers that were totally affordable to a pretty normal working class family.

Do you think it was something like you knew people who had rich kid computers, so you weren't as satisfied with the cheap thrills end of the market?

Koskun
Apr 20, 2004
I worship the ground NinjaPablo walks on
I'll be 44 this year. I built my first computer when I was about 20. It was a duron (667 mhz), with everything but the OS. I remember it being around $600-650.

I built a first gen Ryzen system about 5 years ago I think it was. Everything but the OS and case. It was $725. That was with a 750 watt ps and a 1060 vid card (before the bitcoin bubble).

I remember when I was in 2nd or 3rd grade my Dad bought a computer, as he needed it for work. 486 maybe. I remember it was a grand give or take (he was able to write it off so that helped).


If you shopped and didn't have a need/want for the top of the line setup, prices have been really steady outside of when they first started getting into homes, and not counting the shortage globally now of components of course.

well why not
Feb 10, 2009




https://www.youtube.com/watch?v=nvVafMi0l68

Linus is pumping Right To Repair and has put in 20k from the OnlyFans joke to the cause here:

https://www.gofundme.com/f/lets-get-right-to-repair-passed

gradenko_2000
Oct 5, 2010

HELL SERPENT
Lipstick Apathy
https://twitter.com/HardwareUnboxed/status/1385035841005780992
https://twitter.com/HardwareUnboxed/status/1385051271451660290

It doesn't seem like this is really a dispute between HWU and GamersNexus, so much as a dispute with the audience/community.

I can understand where GN is coming from: running and testing the CPUs under "Intel guidance" means the CPU will actually meet its TDP rating, and if you let it run without power limits, then it might descend into an arms race with all of the other tinkering you can do with AMD short of actual overclocking.

On the other hand, if most motherboards already run without the power limits right out of the box, then the performance measurements captured by GN aren't so "realistic" or "practical". They're consistent against their own testing methodology across all Intel CPUs, but might not match what the user is actually getting unless they go out of their way to re-enable the limits. To GN's credit, if you watch one of their videos reviewing an Intel CPU then you know that the limits exists and might even think of reenabling it, but in this case HWU has a point in that what they're measuring is the out-of-the-box experience.

I guess what HWU is trying to clarify here is that if overclocking a K-SKU CPU voids your warranty, lifting the power limit does not, which might be important to people who ever need to RMA their CPU.

Having said all that, anyone else have an opinion on this difference between how GN and HWU test Intel CPUs? I guess for me the important part is that the viewer is aware of what's being done, so they can calibrate their expectations and comparisons appropriately.

njsykora
Jan 23, 2012

Robots confuse squirrels.


I think it's perfectly fine as long as your methodology is consistent and explained publicly (which GN has done, as well as explaining why they do it like that) so people can actually compare benchmarks across your site knowing everything's playing by the same rules. Running it with whatever it's like out of the box also fine since it's what most people are going to actually use.

Basically I just prefer everyone's consistent in their own benchmarks rather than constantly hunting down some theoretical "perfect" methodology.

mA
Jul 10, 2001
I am the ugly lover.
It's great that Linus (who is arguably the most shilly of techtubers) feels the need to declare that he's" not an anticapitalist activist" in his pro Right to Repair video to reassure the chud section of his fanbase.

mA fucked around with this message at 07:49 on Apr 22, 2021

forest spirit
Apr 6, 2009

Frigate Hetman Sahaidachny
First to Fight Scuttle, First to Fall Sink


to me it seemed like the script was trying to catch every counter-argument against right to repair, not that he was appeasing chudspeople

Whitest Russian
Nov 23, 2013

forest spirit posted:

to me it seemed like the script was trying to catch every counter-argument against right to repair, not that he was appeasing chudspeople

It's nice seeing passionate Linus come out, although the ads take away from the legitimateness of the issue IMO.

FuturePastNow
May 19, 2014


gradenko_2000 posted:

if most motherboards already run without the power limits right out of the box, then the performance measurements captured by GN aren't so "realistic" or "practical". They're consistent against their own testing methodology across all Intel CPUs, but might not match what the user is actually getting unless they go out of their way to re-enable the limits. To GN's credit, if you watch one of their videos reviewing an Intel CPU then you know that the limits exists and might even think of reenabling it, but in this case HWU has a point in that what they're measuring is the out-of-the-box experience.

I think there's a place for both of those methodologies in testing. Be kind of boring if every reviewer did things the same way.

Fame Douglas
Nov 20, 2013

by Fluffdaddy
Different mainboards use different default power limit settings, the only way to have a reasonable comparison between CPUs is to use Intel's (or AMD's) recommendations. Otherwise you're benchmarking some manufacturer's Bios defaults.

Klyith
Aug 3, 2007

GBS Pledge Week

gradenko_2000 posted:

Having said all that, anyone else have an opinion on this difference between how GN and HWU test Intel CPUs? I guess for me the important part is that the viewer is aware of what's being done, so they can calibrate their expectations and comparisons appropriately.

IMO the reasons not to test with whatever settings the motherboard comes with is:

1. Ok, which motherboard do you use as your "testing standard"? Because now mobo settings and power delivery have potential impact on performance. Is it the gold-plated $500 mobo? Or is it the average best-seller mobo that is what the most people are using? Because if the justification for leaving things at OOB mobo settings is "that's the actual user experience", you probably should use a cheaper mobo.

2. It encourages an arms race between motherboard makers. This was a real problem back ~15 years ago when we had a FSB, mobo makers started adding a mhz to the default OOB FSB speed. Then 2 mhz, then 3. And then if you had an unusual or sensitive PCI card in your system, you'd get unstable results.

3. Eventually AMD will join in and let mobo makers ship with PBO options turned on or some poo poo. Which, IMHO, sucks. All this poo poo is adding 1-2% to final CPU performance, at the cost of 50+ watts of power consumption which for most people is a complete waste. And when I say "most people" I'm talking about enthusiast gamers. Maybe I'm just over-sensitive to this issue since loving bitcoin came around.


tldr if Intel wants to have their CPUs score 2% better, they should be consistent with their specs. Call the high-power mode the official spec for the K series, and put the TDP that reflects that spec on the box.

Yaoi Gagarin
Feb 20, 2014

mA posted:

It's great that Linus (who is arguably the most shilly of techtubers) feels the need to declare that he's" not an anticapitalist activist" in his pro Right to Repair video to reassure the chud section of his fanbase.

bourgeois liberals and chuds are not the same thing

njsykora
Jan 23, 2012

Robots confuse squirrels.


I would also put it down to Youtubers' fear of ever being labelled "political" by the algorithm, see also times when GN Steve does an aside about never saying whether or not crypto mining is a bad thing.

Craptacular!
Jul 9, 2001

Fuck the DH

njsykora posted:

I would also put it down to Youtubers' fear of ever being labelled "political" by the algorithm, see also times when GN Steve does an aside about never saying whether or not crypto mining is a bad thing.

I think it's more the audience than the algorithm. Many popular YouTubers in the Linux world go hard on right-wing politics, one even posted his video of attending a Trump rally in the middle of the pandemic. I would say most of those people, even ones I drop for their politics, would support right to repair as a liberty that comes with putting up the cash for purchase. But there's people looking to organize boycotts all over politics, it's just that opensource evangelism is so predominantly white and male that it's no surprise when chuds emerge and there's a strong desire to not lose allies or their knowledge when they do. (It should be noted that actual FOSS projects and organizations are taking a much better ethos on that, creating codes of conduct that is pushing the disgruntled chud factor to the commentator/influencer part of the community, which is part of why this phenomenon occurs.)

To be honest, Linus putting "right to repair" in the spectrum of comerce/communism seems like a bit much. And of course, Linus had his "MONEY!!!!1" video the other month so why would anyone rush to call him a pinko. He's probably reading something he wrote off the cuff, but it's more about anarchy versus just about anything else in the consumer protections space.

The Illusive Man
Mar 27, 2008

~savior of yoomanity~
Who is ReviewTechUSA and does he ever actually review tech or does he just make videos about ~drama~?

Never watched the dude but his videos keep showing up in my feed suggestions.

Purgatory Glory
Feb 20, 2005

Space Racist posted:

Who is ReviewTechUSA and does he ever actually review tech or does he just make videos about ~drama~?

Never watched the dude but his videos keep showing up in my feed suggestions.

He used to be all about reviewing games and talking about how much the gtx 980ti was a better deal than the Titan. But it's evolved to mostly drama talk about the industry or fellow content creators.

Former Human
Oct 15, 2001

Rich the ReviewTech guy had a disagreement with Pat Contri over Diablo Immortal of all things two and a half years ago and still randomly poo poo-talks Pat about it. He's a creep but I guess stirring up drama gets more clicks.

I would blow Dane Cook
Dec 26, 2008
dickbutt

https://www.youtube.com/watch?v=8S-usumuaWo

SalTheBard
Jan 26, 2005

I forgot to post my food for USPOL Thanksgiving but that's okay too!

Fallen Rib

Space Racist posted:

Who is ReviewTechUSA and does he ever actually review tech or does he just make videos about ~drama~?

Never watched the dude but his videos keep showing up in my feed suggestions.

Because of this post I learned that EDP445 is a pedophile.

EDIT: I'm a closet "Youtubers that make videos about other Youtubers fan"

Combat Pretzel
Jun 23, 2004

No, seriously... what kurds?!
https://twitter.com/TheGatorGamer/status/1391988443673251840

:laffo:

--edit:
Apparently this is related to this warning shot bullshit that happened a while ago, after he taunted and dared some other Youtuber douchenozzle to show up at his house.

Combat Pretzel fucked around with this message at 19:55 on May 11, 2021

Klyith
Aug 3, 2007

GBS Pledge Week
How it started:

Klyith posted:

IMO the reasons not to test with whatever settings the motherboard comes with is:

1. Ok, which motherboard do you use as your "testing standard"? Because now mobo settings and power delivery have potential impact on performance.

How it's going:

Intel B560 is a Disaster posted:

Depending on the B560 motherboard, performance of locked 65 watt parts like the 11400 and 11700 can be negatively impacted by over 30%. That’s right, we’re not talking about parts like the Core i9-11900K, but rather processors you will be using with a budget B560 motherboard.
...
Removing the power limits on the B560M Pro with the i7-11700 installed resulted in VRM thermal throttling, and while this still saw the average clock speed raised to 3.8 GHz -- a 900 MHz boost there -- frequent dips to 800 MHz when throttling makes for a horrible experience.
https://www.techspot.com/article/2252-intel-b560-disaster/


So yes this was not at all an academic exercise, Intel's extremely flexible definitions of "TDP", "base clock", and "specification" mean that there are mobos that technically meet Intel's qualifications to run the processors, but cannot support the performance of higher-end mobos.

gradenko_2000
Oct 5, 2010

HELL SERPENT
Lipstick Apathy
My takeaway from that video is:

1. Hardware Unboxed does not actively remove the power limits from their Intel boards - they use out-of-the-box performance, as opposed to Gamers Nexus which (I'm pretty sure) deliberately goes into the board settings to enforce Intel's guidance, regardless of what the board-settings began as. In which case, GN probably wouldn't have found anything of the ordinary from one board to another, even though their approach (again, as they make it clear) technically "leaves performance on the table"

2. You could always run CPUs in VRMs that can't handle it, as in the case of putting a 12-core Zen 2 CPU on an A320, or a 12-core Ryzen 3 CPU on an A520. In either case, the less-capable VRMs are almost certainly going to cause throttling issues, even if you left things "stock", just because AMD likes to do a floating turbo out of the box. I guess the problem with Intel is that a B560 board already isn't assumed to be bottom-of-the-barrel (because it's an H510 that's the lowest tier) and that the 8-core i7-11700 isn't / shouldn't be comparable to a Ryzen 9 3900X.

Chevy Slyme
May 2, 2004

We're Gonna Run.

We're Gonna Crawl.

Kick Down Every Wall.
Lmao at GN steve “I don’t cook my own food”

Pilfered Pallbearers
Aug 2, 2007

K8.0
Feb 26, 2004

Her Majesty's 56th Regiment of Foot
... we see no value ... you're just paying to get ripped off ...

It's... amazing. Worse than I expected.

MonkeyFit
May 13, 2009
I guess I just never realized how scammy and scummy Dell could actually be. I mean, in my defense, I never really took a look at their computers before because I could always build my own.

gradenko_2000
Oct 5, 2010

HELL SERPENT
Lipstick Apathy
https://twitter.com/IanCutress/status/1396430359760359428

what's this all then? I don't want to watch Linus's podcast

avoid doorways
Jun 6, 2010

'twas brillig
Gun Saliva
https://twitter.com/IanCutress/status/1395660056788615172

Linus says he's right but the ship has sailed and no one is going to change their terminology, and apparently this has caused weird nerds to pile on Dr Cutress about it

K8.0
Feb 26, 2004

Her Majesty's 56th Regiment of Foot
A person who isn't a practicing MD but insists on being called "Doctor" is a pedant? The entire world is shocked.

Arivia
Mar 17, 2011

K8.0 posted:

A person who isn't a practicing MD but insists on being called "Doctor" is a pedant? The entire world is shocked.

Ian Cutress has a doctorate in a related field (I forget specifically what), and has been doing good tech journalism at AnandTech for a loooooong time. Dude is legit, he’s earned the title, and he’s trying to get something started to help everyone actually buying technology to get correctly described products.

e: some quick googling says it was awarded for computational chemistry from Oxford in 2011.

Arivia fucked around with this message at 13:55 on May 23, 2021

buffbus
Nov 19, 2012
I listen to the PC Mag podcast and even they use "doctor" when referring to him. I cringe a little every time. The guy is very smart but I can't help but feel this is like an orchestra conductor insisting on being called "maestro" in social situations.

gradenko_2000
Oct 5, 2010

HELL SERPENT
Lipstick Apathy
okay so the dispute is terminology? that when you say RAM is running at "3600", the unit you should be using is "megatransfers per second", and not hertz, because it's actually 1800 megahertz, and then as double-data rate ram, you get one "transfer" on both the rising and falling edge of the wave, so in 1 hertz you get 2 transfers, ergo 1800 MHz = 3600 megatransfers per second

am I understanding this correct?

buffbus posted:

I listen to the PC Mag podcast and even they use "doctor" when referring to him. I cringe a little every time. The guy is very smart but I can't help but feel this is like an orchestra conductor insisting on being called "maestro" in social situations.

DOCTOR JILL BIDEN

K8.0
Feb 26, 2004

Her Majesty's 56th Regiment of Foot

Arivia posted:

Ian Cutress has a doctorate in a related field (I forget specifically what), and has been doing good tech journalism at AnandTech for a loooooong time. Dude is legit, he’s earned the title, and he’s trying to get something started to help everyone actually buying technology to get correctly described products.

e: some quick googling says it was awarded for computational chemistry from Oxford in 2011.

And? I know a very experienced surgeon who now teaches at a pretty good school. She doesn't go by doctor because that's not her principal occupation anymore. I know another woman who is a neuroscience lead researcher and a professor at one of the world's most reputable universities. She doesn't go by doctor either. Only total dinks try to be called "doctor" outside of a medical context, because it's loving irrelevant.

Also, he's "right" but only in one pedantic sense. One can just as easily be pedantic back and argue that while the channel frequency is half-rate, the listed "MHz" number does in fact describe the number of operating cycles per second. Hertz is a proper SI unit, transfers is not! It's unscientific to use MT/s! Also one could be not a moron and recognize that as soon as we made the jump to DDR, we switched to effective frequency because that's what is useful for comparing SDR to DDR speeds. It's a subject worth educating people on because that underlying frequency is sometimes relevant, but the way he's approaching it is pure pedantry.

K8.0 fucked around with this message at 15:19 on May 23, 2021

CoolCab
Apr 17, 2005

glem

gradenko_2000 posted:

okay so the dispute is terminology? that when you say RAM is running at "3600", the unit you should be using is "megatransfers per second", and not hertz, because it's actually 1800 megahertz, and then as double-data rate ram, you get one "transfer" on both the rising and falling edge of the wave, so in 1 hertz you get 2 transfers, ergo 1800 MHz = 3600 megatransfers per second

am I understanding this correct?

think so. honestly it seems like kind of a fair complaint, i've seen several instances of people needing to clarify that their 3600 "mhz" ram will in fact show up in, what CPUZ or ryzen master or whatever people are using, at the non-doubled 1800mhz and that it's not malfunctioning. it's very much an edge case and i generally agree with the perspective that it's a lost battle but i can understand why it might rub some the wrong way, it's confusing.

CoolCab
Apr 17, 2005

glem
i mean my personal crusade against utterly loving stupid and marketing friendly (ie, misleading) technology terminology would probably start with the USB naming standards, how MBps versus Mbps is used to deliberately fool non-technical users which is another thing i've had a friend get confused about, there's no shortage really.

Fame Douglas
Nov 20, 2013

by Fluffdaddy

K8.0 posted:

And? I know a very experienced surgeon who now teaches at a pretty good school. She doesn't go by doctor because that's not her principal occupation anymore. I know another woman who is a neuroscience lead researcher and a professor at one of the world's most reputable universities. She doesn't go by doctor either. Only total dinks try to be called "doctor" outside of a medical context, because it's loving irrelevant.

Who gives a gently caress, imagine getting this angry about people using the titles they earned.

Tuxedo Gin
May 21, 2003

Classy.

I insist that everyone refer to me by my highest level of education, so that everyone I ever meet knows I am a Bachelor.

Perplx
Jun 26, 2004


Best viewed on Orgasma Plasma
Lipstick Apathy
People use MHz and megatransfers interchangeably, they are wrong but I agree the train has left the station long ago and you can't fix it. Its like making hard drive manufacturer's use base 2 terabytes instead of base 10.

Adbot
ADBOT LOVES YOU

njsykora
Jan 23, 2012

Robots confuse squirrels.


Yeah, like it's "technically" wrong but at a point you're just talking in the terms everyone understands, and you only need to explain to someone once about why their RAM shows up at half speed in CPU-Z. Also the performance monitor in Windows and most Bios' account for that double data rate, so you'll see the actual speed in there anyway.

GN's apparently doing a big 'Memory Explained' series soon so I wonder if they'll talk about the Mhz/MT difference there.

  • 1
  • 2
  • 3
  • 4
  • 5
  • Post
  • Reply