2016's $170 GPU vs. 2019's $170 GPUs

[Julio Franco]

Today we're going to review the sub-$200 graphics card market and see how it compares to what we were offered just a few years ago. We recently did this for the $400 price range and it was super interesting. The two most recent additions to this segment are the Radeon 5500 XT and Nvidia's GTX 1650 Super. Let's see how they do against previous generations.

Read the full article here.

https://www.techspot.com/article/1970-then-and-now-170-usd-gpu/

 

[theruck]

All the XTs supposed to be $170 cost $220 so what is the exact meaning of this article?
I just bought a used RX470 and it is overheating like hell so that card had its issues...will end up in trash

 

[Irata]

Looking at the above graphs, it seems like Polaris in its original form was really underappreciated, besting the 1050 Ti on both perf/$ and perf/W.

It also seems like AMD should have offered the 5600XT at the $200 price point in order to have a worthy successor to be equally disruptive.

Then again, I think it's a safe bet that the 1050 series considerably outsold RX 470 / 570, so AMD probably learned from that.

 

[Squid Surprise]

"Lastly something that always comes up is the driver nerfing argument... so we decided to compare the 2013 Tomb Raider results to what we saw in the modern titles. Scaling between the old and newer games looks fairly consistent. The only standout here is the GTX 1650 Super which did better relative to the 5500 XT in Tomb Raider, but this is just a single title, so it’s not surprising to see a result like this. In fact, we’re really surprised how similar the scaling is for the most part.

The R9 370X and GTX 950, for example, were evenly matched in the modern and older games. Same goes for the GTX 1050 Ti and 1650, so we can’t see any driver nerfing going on here."

OK... so you say that there's only 1 older title so even though the 1650 super destroys the 5500, it doesn't really count... But you also claim this means you don't see any "driver nerfing".... you can't have it both ways!

I'd love to see a few more "old" games tested since that's what you're going to be playing a lot of with bottom-tier GPUs anyways...

 

[loki1944]

All the XTs supposed to be $170 cost $220 so what is the exact meaning of this article?
I just bought a used RX470 and it is overheating like hell so that card had its issues...will end up in trash

3x used RX 470s, never had that issue.

 

[loki1944]

"Lastly something that always comes up is the driver nerfing argument... so we decided to compare the 2013 Tomb Raider results to what we saw in the modern titles. Scaling between the old and newer games looks fairly consistent. The only standout here is the GTX 1650 Super which did better relative to the 5500 XT in Tomb Raider, but this is just a single title, so it’s not surprising to see a result like this. In fact, we’re really surprised how similar the scaling is for the most part.

The R9 370X and GTX 950, for example, were evenly matched in the modern and older games. Same goes for the GTX 1050 Ti and 1650, so we can’t see any driver nerfing going on here."

OK... so you say that there's only 1 older title so even though the 1650 super destroys the 5500, it doesn't really count... But you also claim this means you don't see any "driver nerfing".... you can't have it both ways!

I'd love to see a few more "old" games tested since that's what you're going to be playing a lot of with bottom-tier GPUs anyways...

Not true; RX 470 can play modern titles just fine; hell even a 2010 GTX 580 3GB can play quite a few @30FPS.

 

[QuantumPhysics]

During 2016 - 2018 the rise of cryptocurrency caused the price of GPU's to be tremendously inflated. A "Gaming desktop" off the shelf with a 1050Ti was round $1000. While a "Gaming Laptop" with a 1050 was round $1000 - $1500.

Now that cryptocurrency's popularity is mostly dead, prices on the low end have collapsed while the newer, higher-end RTX cards are causing downwards pressure. I just can't see myself taking a GTX over an RTX in 2020.

Furthermore the 4GB of RAM just isn't enough for me.

I run CoD: MW in max detail in 1440p with Ray Tracing on and it visually shows you that it's using almost 6GB / 11GB of the Video RAM available.

 

[Dog8myname]

All the XTs supposed to be $170 cost $220 so what is the exact meaning of this article?
I just bought a used RX470 and it is overheating like hell so that card had its issues...will end up in trash

I had an RX470 for a year or so, no issues at all, and definitely didn't overheat, I'd say you probably got a faulty card. I traded it in (with some other bits and bobs) for a used RX580 2 years ago, which has also been faultless.

 

[Irata]

During 2016 - 2018 the rise of cryptocurrency caused the price of GPU's to be tremendously inflated. A "Gaming desktop" off the shelf with a 1050Ti was round $1000. While a "Gaming Laptop" with a 1050 was round $1000 - $1500.

Now that cryptocurrency's popularity is mostly dead, prices on the low end have collapsed while the newer, higher-end RTX cards are causing downwards pressure. I just can't see myself taking a GTX over an RTX in 2020.

Furthermore the 4GB of RAM just isn't enough for me.

I run CoD: MW in max detail in 1440p with Ray Tracing on and it visually shows you that it's using almost 6GB / 11GB of the Video RAM available.

It really comes down to < $ 200 vs > $ 1,000 for the GPU or alternatively "GPU only" vs. "an entire gaming capable PC".

 

[loki1944]

Got a few R9 Nanos for $125-$148 recently and RX 470 4GBs for $65-$79.

 

[Irata]

Newegg has two RX570 models (Asus ROG Strix and Sapphire Pulse) for $ 109 new (after rebate), including the XBox Game pass and Radeon game bundle (Borderlands 3 or Ghost Recon Breakpoint).

 

[BadThad]

Great article! I'm still trucking along just fine with my $150 RX 480. As you found out, there is no compelling reason for me to upgrade yet. Went from an R9 270 to the RX 480 and I've been quite happy with my game performance.

 

[Nms Barry]

"Lastly something that always comes up is the driver nerfing argument... so we decided to compare the 2013 Tomb Raider results to what we saw in the modern titles. Scaling between the old and newer games looks fairly consistent. The only standout here is the GTX 1650 Super which did better relative to the 5500 XT in Tomb Raider, but this is just a single title, so it’s not surprising to see a result like this. In fact, we’re really surprised how similar the scaling is for the most part.

The R9 370X and GTX 950, for example, were evenly matched in the modern and older games. Same goes for the GTX 1050 Ti and 1650, so we can’t see any driver nerfing going on here."

OK... so you say that there's only 1 older title so even though the 1650 super destroys the 5500, it doesn't really count... But you also claim this means you don't see any "driver nerfing".... you can't have it both ways!

I'd love to see a few more "old" games tested since that's what you're going to be playing a lot of with bottom-tier GPUs anyways...

AMD GPUs perform worse on older and not so AAA titles.
As AMD is constrained on the money they put emphasis on AAA new titles.
If you look at all the titles benchmark of AMD vs Nvidia in 2019, you will get a better idea.

 

[lightsout]

Thank you for adding modern warfare. Haven't seen your other reviews recently but Nvidia seems to have a strong advantage in the title.

 

[krizby]

The sub 200usd GPU landscape would look so much differently had the author choose games that are actually relevant. I would rather look at FPS results for 10 most played game on Steam such as:
Counter-Strike: Global Offensive
Dota 2
PLAYERUNKNOWN'S BATTLEGROUNDS
MONSTER HUNTER: WORLD
Grand Theft Auto V
Tom Clancy's Rainbow Six Siege
Destiny 2
Team Fortress 2
Rust
Football Manager 2020

These games are what people are actually playing for hours and hours, not some flip flop like BFV, COD, Far Cry that EA and Activision **** out every year.
Now my reasoning behind this is that people are buying sub 200usd GPU usually for a single MMO game that they play for a long time. I pretty much only care about PUBG performance, the rest of AAA games I can play at 60fps just fine since the whole game is finished within 30hours usually.

 

[FatZippster]

As an owner of an RX 470 that I got for $155 right before the crypto gpu apocalypse, I am still very happy with my purchase. Years later and it is still a competitive card. Honestly though, I have really mixed feelings about it. On one hand I'm very happy to have a GPU that still holds up well, on the other I'm pretty displeased with both AMD and Nvidia for not making progress in the price range. I'm reaching the end of the operational life of the system I added the RX 470 to and there is nothing in the budget to midrange market segment that would be a worthwhile upgrade. We should have near GTX 1070 performance by now at $200 to $220 and we are nowhere near that. I accept it if tech companies don't want our money, but I can't see how they can stay in business like that.

 

[Nms Barry]

The sub 200usd GPU landscape would look so much differently had the author choose games that are actually relevant. I would rather look at FPS results for 10 most played game on Steam such as:
Counter-Strike: Global Offensive
Dota 2
PLAYERUNKNOWN'S BATTLEGROUNDS
MONSTER HUNTER: WORLD
Grand Theft Auto V
Tom Clancy's Rainbow Six Siege
Destiny 2
Team Fortress 2
Rust
Football Manager 2020

These games are what people are actually playing for hours and hours, not some flip flop like BFV, COD, Far Cry that EA and Activision **** out every year.
Now my reasoning behind this is that people are buying sub 200usd GPU usually for a single MMO game that they play for a long time. I pretty much only care about PUBG performance, the rest of AAA games I can play at 60fps just fine since the whole game is finished within 30hours usually.

Most popular/esports games heavily favor Nvidia.
If benchmarks are done based on most popular games every year AMD will look too bad.

 

[Nms Barry]

As an owner of an RX 470 that I got for $155 right before the crypto gpu apocalypse, I am still very happy with my purchase. Years later and it is still a competitive card. Honestly though, I have really mixed feelings about it. On one hand I'm very happy to have a GPU that still holds up well, on the other I'm pretty displeased with both AMD and Nvidia for not making progress in the price range. I'm reaching the end of the operational life of the system I added the RX 470 to and there is nothing in the budget to midrange market segment that would be a worthwhile upgrade. We should have near GTX 1070 performance by now at $200 to $220 and we are nowhere near that. I accept it if tech companies don't want our money, but I can't see how they can stay in business like that.

yes we are. Gtx 1660 super for $230 offers similar or better performance than gtx 1070.

 

[FordGT40]

I suppose we're looking at the best value cards. I finished a few months ago building my Tower PC, for the monitor I was concerned about gaming graphics running a 4K 34" curved monitor. I had installed two GeForce RTX 2080 Ti Ftw3 Ultra, NVIDIA Turing w/NVLink w/SLI, Overclocked, 2.75 Slot Extreme Cool Triple + iCX2 cooling, 65C Gaming, AI enhanced, RGB, Metal Backplate, 11GB GDDR6. It's fast no matter what I've thrown at it, it keeps up nicely.

 

[Bishstyx]

Overall this is a jaded article that attempts to outline the benefits of the latest GPU offerings as a downfall. If anyone basis any future purchase based on the information here, they must travel back to 2015 and purchase a lower performing predecessor at the same price point. I bothered to create an account here just to describe how pathetic this analysis is.

"The new Radeon 5500 XT was just 31% faster than the RX 470, but interestingly the GTX 1650 Super does much better and is now 52% faster."

Please, someone sell me their RX 470 for $200. Clearly, 31% or 52% performance increases are for morons. TechSpot needs an editor.

 

[AnilD]

If anyone basis any future purchase based on the information here, they must travel back to 2015 and purchase a lower performing predecessor at the same price point.

From the article: "we’re going to review the sub-$200 graphics card market and see how it compares to what we were offered just a few years ago"

Ergo, this is not a review. They did that already. This is a different kind of comparison that is best understood once you read the intro and know what they're trying to show us.

 

[QuantumPhysics]

Does it really make sense to partner a Core i9 9900K with a cheap sub-$200 GPU?

I can't imagine someone with a Core i9 9900K and anything less than a $500 - $1000 card.

My cousin for example: I just built his i9 9900K back in October and even he has a 1080Ti.

These low-end GPUS are more likely to be partnered with low end CPU.

Figure off-the-shelf Core i7.

 

[Julio Franco]

Does it really make sense to partner a Core i9 9900K with a cheap sub-$200 GPU?

This: https://www.techspot.com/article/1637-how-we-test-cpu-gaming-benchmarks/

 

[M8hackr]

You really need to test using a PCIe 4.0 interface. The budget tiers for 500 series chipsets are on the way and PCIe 3.0 completely nerfs the RX 5500. I do agree that AMD should have used an x16 interface, but on PCIe 4.0 I can understand why they wouldn't... Still though, it would make it a more fair comparison since you're not demonstrating the full potential of one single card on your list.

 

[Steve]

Overall this is a jaded article that attempts to outline the benefits of the latest GPU offerings as a downfall. If anyone basis any future purchase based on the information here, they must travel back to 2015 and purchase a lower performing predecessor at the same price point. I bothered to create an account here just to describe how pathetic this analysis is.

"The new Radeon 5500 XT was just 31% faster than the RX 470, but interestingly the GTX 1650 Super does much better and is now 52% faster."

Please, someone sell me their RX 470 for $200. Clearly, 31% or 52% performance increases are for morons. TechSpot needs an editor.

You've somehow missed the entire point of the article, you're also looking at a single result which is a bit odd. If we saw a 52% performance gain on average that would be pretty decent, 31% over 3 years is very meh though.

The real numbers can be seen in the 10 game average graph and you're not looking at 50%+ gains there.

You really need to test using a PCIe 4.0 interface. The budget tiers for 500 series chipsets are on the way and PCIe 3.0 completely nerfs the RX 5500. I do agree that AMD should have used an x16 interface, but on PCIe 4.0 I can understand why they wouldn't... Still though, it would make it a more fair comparison since you're not demonstrating the full potential of one single card on your list.

Correction bud, AMD has nerfed the performance of the 5500 XT, not us! If you're suggesting people need to buy a $150 US+ X570 board to take advantage of the 5500 XT, well then I think this entry-level GPU has bigger issues.