GPUs & CPUs & Enthusiast hardware: Questions, Discussion and fanboy slap-fights - Nvidia & AMD & Intel - Separe but Equal. Intel rides in the back of the bus.

[follow the white rabbit]

How do you cope with needing 1000 watts PSU to run a 1500$ nvidia GPU? It's the worst time of earth to buy anything related to computers.
Just wait,no one wants GPUs that itself melt.

 

[Fcret]

How do you cope with needing 1000 watts PSU to run a 1500$ nvidia GPU? It's the worst time of earth to buy anything related to computers.
Just wait,no one wants GPUs that itself melt.

The 5080 only needs an 850W PSU thoughbeit. The only cards that need 1000 or more right now are the 4090 and 5090.

 

[WelperHelper99]

How do you cope with needing 1000 watts PSU to run a 1500$ nvidia GPU? It's the worst time of earth to buy anything related to computers.
Just wait,no one wants GPUs that itself melt.

I bought a 1000 watt because it was dirt cheap and high quality. Buy a bigger PSU, a 300 watt load should not be a reason to undervolt.

 

[Ibanez RG 350EX]

The real solution is something like a lottery to completely nullify the advantage sweats

If you're trying to buy hardware on launch day, aren't you necessarily a "sweat"? Any normal person will hear about lotteries and scripts to auto-refresh Newegg, say "oh, fuck off" and wait until they don't have to compete for the privilege of spending hundreds of dollars.

 

[George Lucas]

I like GamersNexus but I get annoyed by his lack of holistic reviewing. He focuses too much on raw performance numbers without considering features. I get that it’s hard to test features objectively, but you can’t write off stuff like DLSS and frame gen anymore than you can write off raytracing. It reminds me of when multicore processors first came out and people were skeptical of their performance.

 

[Just Some Other Guy]

I like GamersNexus but I get annoyed by his lack of holistic reviewing. He focuses too much on raw performance numbers without considering features. I get that it’s hard to test features objectively, but you can’t write off stuff like DLSS and frame gen anymore than you can write off raytracing. It reminds me of when multicore processors first came out and people were skeptical of their performance.

It's the most objective way to judge hardware. It's up to buyers to decide how much they value features.

*Edit* Also there are sources that do compare stuff like DLSS/FSR. Just use multiple sources in conjunction.

 

[Gerelatha]

How do you cope with needing 1000 watts PSU to run a 1500$ nvidia GPU? It's the worst time of earth to buy anything related to computers.
Just wait,no one wants GPUs that itself melt.

for someone buying a 90 class GPU a over 1000w PSU isn’t a particularly major expense

 

[DavidS877]

for someone buying a 90 class GPU a over 1000w PSU isn’t a particularly major expense

It's the new electrical panel and new wiring that's the real expense, also the small nuclear reactor in the yard.

 

[George Lucas]

It's the most objective way to judge hardware. It's up to buyers to decide how much they value features.

*Edit* Also there are sources that do compare stuff like DLSS/FSR. Just use multiple sources in conjunction.

I don’t think it is the most objective way though because it’s a fundamentally different approach to what the card is aiming to do. In this environment, AMD is just always going to look better in those price/performance numbers, but I don’t really think AMD is delivering the better product (Nvidia Gen 5 is looking like a disaster however, but that’s besides the point).

I just think Steve could be more forward-thinking in trying to find a way to quantify the features as he’s proven himself to be trustworthy and could act more like a leader in this space. It’s just a minor nitpick. I still like his reviews.

 

[LegoTugboat]

It's the new electrical panel and new wiring that's the real expense, also the small nuclear reactor in the yard.

Nonsense. All the talk about nuclear reactors, completely silly.

A diesel generator's more than capable of handling a modern GPU. You simply have to keep it filled up, and be able to tolerate the noise. And ideally not have neighbours who will complain about said noise.

 

[Prehistoric Jazz]

 

[The Ugly One]

It's because devs had to use a hybrid approach off adding RT over raster. Now we're starting to see RT only games and they are resonable(Doom the Dark ages asks for an RTX 2060 for 1080p@60). You sound like every gamer that rages about new tech that becomes standard.

DtDA isn't a RT-only game, or it wouldn't be coming to the PS5. The reason RT performs badly is tracing one ray per pixel and bouncing it around a scene is far more expensive than calculating a pixel color based solely on the polygon at that location.

 

[The Ugly One]

It's the most objective way to judge hardware. It's up to buyers to decide how much they value features.

Reviewing products without reviewing key features because competitors don't have them isn't "objective." Imagine if an auto reviewer was comparing two pickup trucks and simply didn't talk at all about the extended cab in one of them because the other truck didn't have that as an option.

At the minimum, reviewers should be reviewing games with and without both DLSS and FSR and give some kind of qualitative assessment. Like the fact DLSS can reconstruct in-world text in Helldivers II, but FSR can't, is kind of a big deal that ruins the playability of the game with FSR.

There are ways to review things that are hard to quantify, double-blind studies being one of the best. A double-blind study would be a great way to review upscaling technology.

 

[Just Some Other Guy]

double-blind studies being one of the best. A double-blind study would be a great way to review upscaling technology.

How would they relay that information to the viewer, though? You either have to trust their opinion on it or hope that what they show you on screen is an accurate representation of what you would see in person....which it probably isn't going to be.

Digital Foundry is the source everyone usually goes to for this, and even then you have to trust what they're saying. People sometimes have different experiences because the "quality of upscaling" is completely subjective. What FPS people get with what hardware and other pc specs is not (or at least is a lot less subjective, someone's PC can be more or less out of whack and give somewhat different results).

The size of an extended cab is objective information. You can even come up with metrics of measurement to make it easier for people to visualize what they can fit (how many luggages of X size, for instance).

*Edit* I guess if you went above and beyond to get some truly average people to do the test, or just get some guys the majority of people trust that could be an okay-ish way.

 

[4090 Chad]

Shit like this should be punished like pedos… millstone around the neck and thrown into deep water

 

[Just Some Other Guy]

View attachment 7067771

Shit like this should be punished like pedos… millstone around the neck and thrown into deep water

I recall seeing a FB listing and immediately recognized the apartment buildings across from the nearest Micro Center through his windshield lol.

This guy somehow 1 upped it.

 

[WULULULULU]

View attachment 7067771

Shit like this should be punished like pedos… millstone around the neck and thrown into deep water

lol. You can have it, ASUS is a shit brand anyway. My target is Sapphire Pulse and they'll have steady supply in the upcoming months.

 

[Get the rope Macaulay!]

Just checked Microcenter, and literally every store outside of the ones on the west coast and illinois have at least one 9070 in stock, some places like ohio has like 20 of them and a couple XTs. Honestly crazy to me that scalpers even exist for a product that isn't sold out,

and most console devs preferring to focus on cutting edge visuals instead of heavy physics simulation.

which is an odd take because the Arkham series still has some of the best visuals, and does PhysX, like i'd assume that would just be where graphics would be in the 2010s and they could then dedicate their time to the other shit.

I saw the images of hundreds in stock, but those would certainly be limited to the largest population centers/biggest outlets.

not really just live by a microcenter.

ASUS is a shit brand anyway.

How can someone besmirch the Republic of Gamers?

 

[George Lucas]

How would they relay that information to the viewer, though? You either have to trust their opinion on it or hope that what they show you on screen is an accurate representation of what you would see in person....which it probably isn't going to be.

Digital Foundry is the source everyone usually goes to for this, and even then you have to trust what they're saying. People sometimes have different experiences because the "quality of upscaling" is completely subjective. What FPS people get with what hardware and other pc specs is not (or at least is a lot less subjective, someone's PC can be more or less out of whack and give somewhat different results).

The size of an extended cab is objective information. You can even come up with metrics of measurement to make it easier for people to visualize what they can fit (how many luggages of X size, for instance).

*Edit* I guess if you went above and beyond to get some truly average people to do the test, or just get some guys the majority of people trust that could be an okay-ish way.

That’s the whole point of a double blind: you don’t know what you’re looking at so you can’t introduce bias. The problem is that people are so stubborn that they don’t accept the results. You see this everywhere in the audiophile community.

 

[Just Some Other Guy]

That’s the whole point of a double blind: you don’t know what you’re looking at so you can’t introduce bias. The problem is that people are so stubborn that they don’t accept the results. You see this everywhere in the audiophile community.

Maybe they don't accept the results because they honestly see/hear/taste it differently than the people who did the test? If a double blind test has people saying "Pepsi tastes better than coke" and then I do the same thing but like the coke more does that mean I'm being stubborn?

It's blind to the people doing the actual test/review. The end user might have a different experience.