Started By
Message

re: Finch's Guide to Obtaining a GPU/CPU and Trading Thread

Posted on 7/21/22 at 9:05 pm to
Posted by SaintEB
Member since Jul 2008
22776 posts
Posted on 7/21/22 at 9:05 pm to
210 MHz when utilization is 0%.
Posted by Joshjrn
Baton Rouge
Member since Dec 2008
27141 posts
Posted on 7/21/22 at 10:13 pm to
No, I mean what’s your clock when you are at the early stages of being under load. 99% utilization but just around 55c or 60c.
Posted by SaintEB
Member since Jul 2008
22776 posts
Posted on 7/22/22 at 8:58 am to
Sorry. I'll have to check that when I get home.
Posted by Joshjrn
Baton Rouge
Member since Dec 2008
27141 posts
Posted on 7/22/22 at 9:06 am to
No worries, man. Just trying to get an idea of what kind of juice is left to squeeze
Posted by SaintEB
Member since Jul 2008
22776 posts
Posted on 7/22/22 at 9:36 pm to
quote:

99% utilization but just around 55c or 60c.


Does this make sense?

90-91% utilization - 60c - 1935-1950 MHz
100% utilization - 65c - 1815 MHz
Posted by Joshjrn
Baton Rouge
Member since Dec 2008
27141 posts
Posted on 7/22/22 at 10:00 pm to
quote:

Does this make sense? 90-91% utilization - 60c - 1935-1950 MHz 100% utilization - 65c - 1815 MHz


Sounds like we are dealing with apples to oranges workloads. That’s part of why synthetic benchmarks are good for this kind of thing. It’s the exact same workload each time.

Have you downloaded EZbench yet?

Eta: Everyone should go buy 3DMark next time it’s on sale on Steam :hmph:
This post was edited on 7/22/22 at 10:02 pm
Posted by SaintEB
Member since Jul 2008
22776 posts
Posted on 7/23/22 at 1:47 pm to
Yes. I have EZBench.
Posted by Joshjrn
Baton Rouge
Member since Dec 2008
27141 posts
Posted on 7/23/22 at 2:41 pm to
quote:

Yes. I have EZBench.


Cool. Run it a couple of times and tell me your clock speed about ten seconds in (there’s usually some early frickery, just take what you would call normal) and the clock speed near the end when your temp has gone up a bit. It’s not going to be as good as Port Royal, but it will do.

Actually, turn on looping and let it run a few times until you have temp soaked, and tell me that clock speed.

Eta: If you’re on discord, I’m happy to walk you through all of this there as well.
This post was edited on 7/23/22 at 2:43 pm
Posted by SaintEB
Member since Jul 2008
22776 posts
Posted on 7/24/22 at 12:32 am to
I am on discord but I rarely use it.

EB#6690

I can dust it off to move this there.
Posted by Joshjrn
Baton Rouge
Member since Dec 2008
27141 posts
Posted on 7/24/22 at 7:59 am to
It’s of no moment to me. Just depends on your level of patience
Posted by SaintEB
Member since Jul 2008
22776 posts
Posted on 7/24/22 at 9:47 pm to
quote:


It’s of no moment to me. Just depends on your level of patience


Understood. We probably should as we have taken over the thread with for a page or so with this.
Posted by finchmeister08
Member since Mar 2011
35760 posts
Posted on 8/5/22 at 8:23 pm to
3090 Ti for $1150 anyone?

LINK
Posted by SaintEB
Member since Jul 2008
22776 posts
Posted on 8/6/22 at 12:10 am to
There are a couple of youtubers having a heart attack about Newegg starting a justgpu.com to sell all the 30 series in stock. They are claiming that the benchmarks on the site are false.

Gamers Nexus is the channel. The guy, Steve, seems to know what he is talking about but is definitely a component snob. Probably sponsored stuff "is better".

JayzTwoCents is pushing hard for people to buy cards right now. Some others called him out for doing it.

Both of these guys hate Newegg and don't recommend buying from them, but who has a better selection for places that don't have Micro Centers?
This post was edited on 8/6/22 at 11:47 am
Posted by Richard Dangler
Huntsville, AL
Member since Dec 2015
725 posts
Posted on 8/7/22 at 8:18 pm to
If you have a local bestbuy then just keep an eye on what they have in stock.
Posted by Joshjrn
Baton Rouge
Member since Dec 2008
27141 posts
Posted on 8/7/22 at 8:59 pm to
quote:

There are a couple of youtubers having a heart attack about Newegg starting a justgpu.com to sell all the 30 series in stock. They are claiming that the benchmarks on the site are false.


The frame rates being quoted are non-verifiable at best and false at worst, yes.

quote:

Gamers Nexus is the channel. The guy, Steve, seems to know what he is talking about but is definitely a component snob. Probably sponsored stuff "is better".


Steve is one of the most thorough, most unbiased sources in the space. He gives zero fricks about sponsors.

quote:

JayzTwoCents is pushing hard for people to buy cards right now. Some others called him out for doing it.


I think he’s over correcting from last Gen, but I understand where he’s coming from. He, and everyone else, told people not to buy because RTX 3000 was about to drop. And then RTX 2000 cards went from collecting dust on store shelves to impossible to buy for anything near retail two years ago once it did.

The advice I’ve been giving people is that if you’re looking for a high end vanity upgrade, hold off. But if your card has been limping along, go buy a nice middle of the road 3070 or something like that while you know you can. Because there is no guarantee you’ll be able to immediately get your hands on a next Gen card anytime soon.

quote:

Both of these guys hate Newegg and don't recommend buying from them, but who has a better selection for places that don't have Micro Centers?


No one does. They are telling you to avoid them anyway because of years of shady business practices. But it’s not like they are going to pillory you if you do. They just don’t want to hear whining from anyone if Newegg fricks them
Posted by SaintEB
Member since Jul 2008
22776 posts
Posted on 8/8/22 at 8:13 am to
quote:

Steve is one of the most thorough, most unbiased sources in the space. He gives zero fricks about sponsors.



This is good to know. He seems to have a large following.

quote:

I think he’s over correcting from last Gen, but I understand where he’s coming from.


I think the 4000 series launch will be a shite show like all other popular tech. There are just too many unknowns. That's why I pulled the trigger on the 3080FE when given a chance.



Posted by Joshjrn
Baton Rouge
Member since Dec 2008
27141 posts
Posted on 8/8/22 at 9:17 am to
quote:

This is good to know. He seems to have a large following.


He started out as a very niche, very high level (re: technically sophistication required) tech youtuber. Over the years, he has invested tens of thousands of dollars, if not more, into testing equipment so that his "reviews" can be as objective as possible. In the last year or two, he has started a push into legitimate tech investigative journalism, which is sorely needed in the space. Tech Jesus isn't everyone's cup of tea, because he can get so in depth that it can get boring, but Steve is one of the most respected people in the entire tech community.

Now, that's not to say that he's perfect, because he certainly isn't. He has a bias (for lack of a better description) towards performance testing that can be perfectly replicable apples to apples, which I find sometimes loses the forest for the trees, but more often than not, his testing results are going to be the best around.

And to circle back to sponsorship, the overwhelming majority of his funding comes from his store and his Patreon supporters. And as a rule, his actual "sponsors" don't produce tech he reviews, and are usually more tech-adjacent, like Squarespace.
Posted by SaintEB
Member since Jul 2008
22776 posts
Posted on 8/8/22 at 9:30 am to
quote:

He has a bias (for lack of a better description) towards performance testing that can be perfectly replicable apples to apples, which I find sometimes loses the forest for the trees, but more often than not, his testing results are going to be the best around.


I will tell you this, I watched probably 5-6 case reviews from him and he uses the exact same rig in every case. He calls back to previous editions of the same case, at times, and criticisms that were changed (for better, in his opinion). My purchase of the NZXT H7 Flow was largely due to his, and other reviewers, comments about its performance but also how NZXT's changes made them perform better.
Posted by Joshjrn
Baton Rouge
Member since Dec 2008
27141 posts
Posted on 8/8/22 at 10:13 am to
quote:

I will tell you this, I watched probably 5-6 case reviews from him and he uses the exact same rig in every case. He calls back to previous editions of the same case, at times, and criticisms that were changed (for better, in his opinion). My purchase of the NZXT H7 Flow was largely due to his, and other reviewers, comments about its performance but also how NZXT's changes made them perform better.


Yep. He's utterly performance focused. Which for most people is absolutely what they need. When it gets to the extremes of the hobby, especially when it edges more towards vanity, his numbers aren't always useful. To stick with cases, I own an o11xl. According to Steve, the thermal performance of that case is utterly mediocre. But the testing methodology is using his "standard case testing methodology" which uses three fricking fans. So sure, do other cases perform better with three fans? Of course. This case is god damned huge, and three fans isn't sufficient to replace all the air. What I want to know is how it will perform if I put nine, or hell, just six, fans in it. Literally no one is paying $200 for an o11xl and putting three fans in it
Posted by MetroAtlantaGatorFan
Member since Jun 2017
15598 posts
Posted on 8/8/22 at 3:07 pm to
quote:

3090 Ti for $1150 anyone?

It's weird how the high end cards are good deals now but the lower end is still overpriced.
first pageprev pagePage 66 of 67Next pagelast page

Back to top
logoFollow TigerDroppings for LSU Football News
Follow us on Twitter, Facebook and Instagram to get the latest updates on LSU Football and Recruiting.

FacebookTwitterInstagram