Basic RTX 2070 Super vs Top RX 5700 XT?

[Slazyel]

Hi guys, currently where i live there's 3 cards i'm interested. Those are they:

 

Sapphire RX 5700 XT Pulse: $515

Powercolor RX 5700 XT Red Devil: $560 

GalaxRTX 2070 Super EX 1click oc: $630

 

Also, amd is giving AAA games with the purchase, nvidia dont.

 

I iam considering the Red Devil, its $45 more than the pulse and $70 less than the most basic 2070s.

 

What you guys think?

[Trystanrak]

It really depends on the games that you are playing really. I found I got a ryzen processor which was awesome so I went to try a Rx graphics card as well. What I found is that in the games I played there was some frame stuttering on the Rx graphics card that I didn't have on my older nvidea. Ended up giving my brand new Rx 580(a year ago or so) to my wife and getting a gtx 1080 ftw2. Honestly though on a general level I find nvidea performs a little better and more consistent but if all you are playing are amd preferred games then go amd lol

[Tom_nerd]

I would definitely recommend the 2070 super or if that is above your budget just go for a 2070 (non-super) because of the superior features like RTX, H.264 encoding, and Nvenc encoder engine but if you don`t need any of those then you might be the only one that should buy a 5700 XT (joking)

[Juular]

Almost had a heart attack looking at these prices, Canada i guess ?

Choosing between Pulse and Red Devil, it's simply the cheapest one. Whether it worth going for RTX2070 Super, with this price difference probably not, but be aware that AMD drivers are mess.

[Trystanrak]

1 minute ago, Juular said:

Almost had a heart attack looking at these prices, Canada i guess ?

Choosing between Pulse and Red Devil, it's simply the cheapest one. Whether it worth going for RTX2070 Super, with this price difference probably not, but be aware that AMD drivers are mess.

Ya it sounds to be Canada lmao.... It's great with our great Canadian shmeckles lol

[Jurrunio]

2070S is better in every way, I'd get that as long as you can afford it. The Galax card is not bad.

[Slazyel]

2 minutes ago, Juular said:

Almost had a heart attack looking at these prices, Canada i guess ?

Choosing between Pulse and Red Devil, it's simply the cheapest one. Whether it worth going for RTX2070 Super, with this price difference probably not, but be aware that AMD drivers are mess.

Worst, Brazil, haha.

[Jeppes]

11 minutes ago, Trystanrak said:

It really depends on the games that you are playing really. I found I got a ryzen processor which was awesome so I went to try a Rx graphics card as well. What I found is that in the games I played there was some frame stuttering on the Rx graphics card that I didn't have on my older nvidea. Ended up giving my brand new Rx 580(a year ago or so) to my wife and getting a gtx 1080 ftw2. Honestly though on a general level I find nvidea performs a little better and more consistent but if all you are playing are amd preferred games then go amd lol

Swapping to a much faster gpu tends to give better performance.

[Trystanrak]

4 minutes ago, Jeppes said:

Swapping to a much faster gpu tends to give better performance.

Most of the time for sure. however sometimes I have found that my friends comp that we just built has an Rx 590 and performs a bit better on some titles against my 1080. Always check the titles you play is all I'm saying lol it seems like amd is a bit better when playing red dead for example so if all you play is red dead why pay the extra 150 ish bucks to get the rtx

[Jeppes]

1 minute ago, Trystanrak said:

Most of the time for sure. however sometimes I have found that my friends comp that we just built has an Rx 590 and performs a bit better on some titles. Always check the titles you play is all I'm saying lol it seems like amd is a bit better when playing red dead for example so if all you play is red dead why pay the extra 150 ish bucks to get the rtx

Hard to believe. Even 1080fe should be about 45% faster than rx590. Most likely something bottlenecking your system(capped fps etc) or you run different resolutions or settings.

 

Hwunboxed usually tests many games so check the lists and pick a card based on what you play. If you only play competitive fps:s on low settings, I would consider between 2060 and 5600XT.

[Stu_Bear]

Get the 2070 super and sleep well at night..

[Trystanrak]

5 minutes ago, Jeppes said:

Hard to believe. Even 1080fe should be about 45% faster than rx590. Most likely something bottlenecking your system(capped fps etc) or you run different resolutions or settings.

 

Hwunboxed usually tests many games so check the lists and pick a card based on what you play. If you only play competitive fps:s on low settings, I would consider between 2060 and 5600XT.

Trust me I know I don't understand either but literally every piece of my build is better than his except that he is running everything on an m.2 vs I am a scrub and still using HDD for storage os on SSD but that should not affect gaming performance... Life is hard lol

[Constantin]

Pick the RTX 2070 Super, to have good bight sleeps

[Rotta]

Red Devil RX 5700 XT owner here, dont buy this GPU. Drivers are so bad that even me as an AMD "fan" wanted to return this pile of doo, unfortunately for me its too late.

[BTGbullseye]

7 hours ago, Tom_nerd said:

because of the superior features like RTX, H.264 encoding, and Nvenc encoder

The encoding on the 2070 Super is not noticeably better quality than the one on the 5700 XT, it's just a much older implementation so it has more software that supports it.

6 hours ago, Rotta said:

Red Devil RX 5700 XT owner here, dont buy this GPU. Drivers are so bad that even me as an AMD "fan" wanted to return this pile of doo, unfortunately for me its too late.

You sound like an Nvidia shill, not someone who has actually done at least basic troubleshooting on their system.

[Rotta]

5 hours ago, BTGbullseye said:

The encoding on the 2070 Super is not noticeably better quality than the one on the 5700 XT, it's just a much older implementation so it has more software that supports it.

You sound like an Nvidia shill, not someone who has actually done at least basic troubleshooting on their system.

DDU'd drivers, used every driver from 19.12.1 to 20.2.1, all terrible, also tested the 19.11.x drivers, no luck. Tell me what I need to do to ""troubleshoot"", ill do everything I can but I have come to the conclusion that its literally unfixable. A normal consumer should never have to troubleshoot their own system. The older drivers are old aswell, so I would have to sacrifice performance if I used them.

 

The ease of use with AMD cards isnt there, when with NVIDIA I have heard almost no complaints. Last time I owned an NVIDIA card was a GT 520 or something.

[BTGbullseye]

9 hours ago, Rotta said:

I have come to the conclusion that its literally unfixable.

Then RMA the card. If changing the drivers didn't fix it, then it's most likely NOT the drivers at fault.

9 hours ago, Rotta said:

A normal consumer should never have to troubleshoot their own system.

That has never not been a thing in the PC world... Unless you're a fan of taking your entire system in to the Genius Bar for replacement, every user SHOULD be troubleshooting their systems when reality hits and something goes wrong.

9 hours ago, Rotta said:

The older drivers are old aswell, so I would have to sacrifice performance if I used them.

Not really, unless you're talking about specific games. I still have yet to get better performance than I did with 19.11.1. (though I much prefer the increased stability on my overclocks that I get with the newer driver)

9 hours ago, Rotta said:

The ease of use with AMD cards isnt there, when with NVIDIA I have heard almost no complaints. Last time I owned an NVIDIA card was a GT 520 or something.

So... You haven't even touched their hardware/software for Nvidia cards newer than 9 years old... And you think people should actually listen to you spouting this BS?

[Zberg]

I personally have had great experience with the gigabyte cards and they tend to sell at some of the lowest price points.

 

I have been harping on this for a while, but I would take a cheap entry level 2070S over any of the 5700XT's, hands down.  The windforce 3 fan is right around MSRP/reference and the gaming OC version is like 20 bucks more. 

 

AMD has cracked the code on CPU's and dominated the value/performance/reliability scene.  GPU division hasnt gotten that memo yet it.

 

 

[Rotta]

53 minutes ago, BTGbullseye said:

Then RMA the card. If changing the drivers didn't fix it, then it's most likely NOT the drivers at fault.

That has never not been a thing in the PC world... Unless you're a fan of taking your entire system in to the Genius Bar for replacement, every user SHOULD be troubleshooting their systems when reality hits and something goes wrong.

Not really, unless you're talking about specific games. I still have yet to get better performance than I did with 19.11.1. (though I much prefer the increased stability on my overclocks that I get with the newer driver)

So... You haven't even touched their hardware/software for Nvidia cards newer than 9 years old... And you think people should actually listen to you spouting this BS?

A 45k person poll disagrees with you and so do alot of reviewers. 

If you are saying its the GPU's fault, why do like 25% of the card owners have a broken card? Makes no sense. AMD would never push that.

A lot of people say they have problems with their RX cards and you call me about bullshit? Holy fuck. "Spouting BS" btw

[BTGbullseye]

3 hours ago, Rotta said:

A 45k person poll disagrees with you and so do alot of reviewers.

Not based on actual numbers of real issues happening means that this is useless info.

3 hours ago, Rotta said:

If you are saying its the GPU's fault, why do like 25% of the card owners have a broken card? Makes no sense. AMD would never push that.

Prove those numbers please, because I have heard no one else claim that 1-in-4 users even have issues, much less a broken card.

3 hours ago, Rotta said:

A lot of people say they have problems with their RX cards and you call me about bullshit? Holy fuck. "Spouting BS" btw

A lot of people are having problems with their RTX cards too... Just look at the topics in this forum for proof of that. You need to stop with this hatred, it'll eat you up inside.

[Rotta]

6 minutes ago, BTGbullseye said:

Not based on actual numbers of real issues happening means that this is useless info.

Prove those numbers please, because I have heard no one else claim that 1-in-4 users even have issues, much less a broken card.

A lot of people are having problems with their RTX cards too... Just look at the topics in this forum for proof of that. You need to stop with this hatred, it'll eat you up inside.

Ill RMA my card and I will post my results.

 

"Prove those numbers" Why'd you need numbers specifically? The poll literally gives you direction on which drivers are better and what not.

I said 25% giving benefit of the doubt that it isnt 50% like the study says. AMD has knowingly had problems with drivers SINCE AGES.

[BTGbullseye]

7 minutes ago, Rotta said:

"Prove those numbers" Why'd you need numbers specifically? The poll literally gives you direction on which drivers are better and what not.

I said 25% giving benefit of the doubt that it isnt 50% like the study says. AMD has knowingly had problems with drivers SINCE AGES.

The number of people voting in a Hardware Unboxed poll is significantly fewer than the number of people actually using the cards, not to mention there are TONS of poll voting bots out there designed specifically to skew poll results. Polls rarely if ever actually represent reality anymore.

 

Heck, there's even the obvious of that people looking for help with problems are vastly more likely to find the poll than people that aren't looking to fix something.

[Rotta]

11 minutes ago, BTGbullseye said:

The number of people voting in a Hardware Unboxed poll is significantly fewer than the number of people actually using the cards, not to mention there are TONS of poll voting bots out there designed specifically to skew poll results. Polls rarely if ever actually represent reality anymore.

 

Heck, there's even the obvious of that people looking for help with problems are vastly more likely to find the poll than people that aren't looking to fix something.

I highly doubt it considering he has 455K subs and 45K voted on that one, but I can agree with people looking for help with problems, but the numbers have a huge difference.

[BTGbullseye]

Just now, Rotta said:

I highly doubt it considering he has 455K subs and 45K voted on that one, but I can agree with people looking for help with problems, but the numbers have a huge difference.

Do you really think 45k people around the world (with the probability of bots skewing the vote) is representative of the millions of people actually using the GPUs? (just look at the Steam Survey page to see how many people you're applying this to)

[Rotta]

2 minutes ago, BTGbullseye said:

Do you really think 45k people around the world (with the probability of bots skewing the vote) is representative of the millions of people actually using the GPUs? (just look at the Steam Survey page to see how many people you're applying this to)

You are the type of person to want EXACTLY EVERYTHING or its incorrect. Prove its botted please. Numbers against nothing.

If it doesnt fit with what you think you cant accept it.