[SOLVED] New RTX 2060 heavily under performing

Oct 29, 2019
3
0
10
Hello everyone,

Before i start this is my current desktop specs:
GPU: Msi ventus RTX 2060
CPU: Ryzen 5 1600
MB: MSI Tomahawk b350
Ram: Single 8 gb ddr4 ram stick @2400mhz
PSU: Evga 550W B3
Storage: 256GB SSD + 1TB HDD

Yesterday I received my new GPU and i installed it into my pc straight away. But when I went to test the fps with some games I was really underwhelmed. My GPU could barely hit 60 fps in games like battlefield 1, sea of thieves, call of duty modern warfere on ultra settings. Some times it the fps would jump to 80-100 fps but this was rare, most of the times it stayed between 50-60 fps. I also tried to play some games on low settings but the fps didnt increase much from that, I get 80-90 fps in battlefield 1 on low settings. My previous GPU was a 1050Ti and I would assume that a RTX 2060 would run much better than that, but right now it seems like this is not the case. Am I not supossed to get much more fps with the current GPU i have (like 100fps+ in games like battlefield 1 on ultra settings)? Did I maybe receive a faulty GPU? The fps I'm getting is just not right for a RTX 2060, if anyone has any ideas on this matter please do share. Thank you.
 
Solution
The 2060 is a mid range card.
You frame rate is about right on ultra settings, probably a little low because of the memory handicap you have.
Slow single channel ram hinders the CPU performance quite a bit.
Hello everyone,

Before i start this is my current desktop specs:
GPU: Msi ventus RTX 2060
CPU: Ryzen 5 1600
MB: MSI Tomahawk b350
Ram: Single 8 gb ddr4 ram stick @2400mhz
PSU: Evga 550W B3
Storage: 256GB SSD + 1TB HDD

Yesterday I received my new GPU and i installed it into my pc straight away. But when I went to test the fps with some games I was really underwhelmed. My GPU could barely hit 60 fps in games like battlefield 1, sea of thieves, call of duty modern warfere on ultra settings. Some times it the fps would jump to 80-100 fps but this was rare, most of the times it stayed between 50-60 fps. I also tried to play some games on low settings but the fps didnt increase much from that, I get 80-90 fps in battlefield 1 on low settings. My previous GPU was a 1050Ti and I would assume that a RTX 2060 would run much better than that, but right now it seems like this is not the case. Am I not supossed to get much more fps with the current GPU i have (like 100fps+ in games like battlefield 1 on ultra settings)? Did I maybe receive a faulty GPU? The fps I'm getting is just not right for a RTX 2060, if anyone has any ideas on this matter please do share. Thank you.


like Unolocogringo has said your lowering your performance with a single stick of ram.

this is the ram hit i use


name of ram
TLGD416G2400HC14DC01
https://www.amazon.co.uk/gp/product/B01MRRL7X5/ref=ppx_yo_dt_b_search_asin_title?ie=UTF8&psc=1

i can vouch that this ram works in msi motherboards as i use this on my own 1600 ryzen
its also on the support site for your board for msi

duel channel matters on ryzen and especially the first gen ryzen are very ram sensitive.
 
RTX2060 is 2x more faster than 1050TI, so you're assuming right about the difference. It's huge and should be huge.

There's a huge mistake on this setup. You're running on Single Channel and Dual Channel is mandatory for today games. Look at this video below and notice the HUGE difference between single channel and dual channel.

i'm not that good with maths, but we're talking abou 70% less performance on Single Channel.

View: https://youtu.be/-k5wA7EFwpo?t=83
 
  1. As Mentioned Above Dual Channel Is a must for modern games .
  2. The reccomended Ram Size for RTX 2060 Is 16Gb so try going with a 2x8 Ram Upgrade . I did it and it made a huge differnece i also have rtx 2060 and it is a great card it easily Goes Higher than 80 fps in all the games you mentioned on very high/ ultra settibgs
 
Battlefield 1/3/4/5 all have higher fps with faster ram with tighter timings and dual channel mode. You would want to get at least a 2x8GB 3000-3200Mhz kit with CL14-15 timings to gain 20-25fps over 2400mhz.
Dual Channel is the problem. Better memory (higher clock + lower latency) = better frames on any modern game, but the problem here is because Single Channel Mode.
 
Dual Channel is the problem. Better memory (higher clock + lower latency) = better frames on any modern game, but the problem here is because Single Channel Mode.
All the Battlefield games using the Frostbite engine all gain higher fps from faster memory, not just using dual channel mode. If the OP were to buy a new memory kit, it would just make sense to get a 3000/3200 kit and gain all the benefits of faster ram and dual channel mode. The fps increase could potentially be as much as 90-95% with a 3200mhz kit in dual channel mode.

View: https://www.youtube.com/watch?v=XeQMVhMkTm0
 
All the Battlefield games using the Frostbite engine all gain higher fps from faster memory, not just using dual channel mode. If the OP were to buy a new memory kit, it would just make sense to get a 3000/3200 kit and gain all the benefits of faster ram and dual channel mode. The fps increase could potentially be as much as 90-95% with a 3200mhz kit in dual channel mode.

View: https://www.youtube.com/watch?v=XeQMVhMkTm0
Agreed. But if he choose to buy another 2400Mhz, he will have already a massive boost on FPS. A better kit would make these gains even higher, we both agree, but it's not necessary to fix the issue.
 
Agreed. But if he choose to buy another 2400Mhz, he will have already a massive boost on FPS. A better kit would make these gains even higher, we both agree, but it's not necessary to fix the issue.

I don't see a point in mentioning this at all.

Mixing ram sticks is not recommended and could bring potential unstabilty to his system. If he where to get something it should be 2x matched sticks.
 
I don't see a point in mentioning this at all.

Mixing ram sticks is not recommended and could bring potential unstabilty to his system. If he where to get something it should be 2x matched sticks.
I work with it in the free time to extra money. 1 to 3 PCs for week. This is much more rare than people are aware. I have this just one time from a year to now. And I always run some good tests before send back the PCs.

It's not a popular statement or the one that industry would like to make people belive, but it's the truth.

For OC is a nightmare, but stock generally it works fine.

Anyway, if you buy a memory stick that eventually doesnt work with stick that you have already, you just have to return it. I don't know if you know, but in some countries, memory stick is incredible expensive (due to taxes) and not everyone in the world have payments in dollar or euro.
 
I work with it in the free time to extra money. 1 to 3 PCs for week. This is much more rare than people are aware. I have this just one time from a year to now. And I always run some good tests before send back the PCs.

It's not a popular statement or the one that industry would like to make people belive, but it's the truth.

For OC is a nightmare, but stock generally it works fine.

Anyway, if you buy a memory stick that eventually doesnt work with stick that you have already, you just have to return it. I don't know if you know, but in some countries, memory stick is incredible expensive (due to taxes) and not everyone in the world have payments in dollar or euro.


Still suggesting that to someone that does not do that for a living or in they free time is not a good idea.
This user might start having issues with his PC right away, maybe not or maybe later.

Them not knowing that 2 different ram sticks can cause issues will start to troubleshoot and before coming to a solution the return window of the item might run out.

Then all of a sudden you are stuck with a ram stick that you can not use and will have to buy 2 new ones.

It's not about what you do in you free time with building/repairing PC and testing them after. It's about what you should be recommending to a general consumer that does not know better and what would be better for them in the long run.
 
This user might start having issues with his PC right away, maybe not or maybe later.
This is the only reason why we run benchmark, Prime, Aida, memtest and other programs. To be sure that people will haven't any problems later.

Anyway, You just follow what manufactors says. You're not wrong at all. Let's finish this here, it's better.
 
This is the only reason why we run benchmark, Prime, Aida, memtest and other programs. To be sure that people will haven't any problems later.

Anyway, You just follow what manufactors says. You're not wrong at all. Let's finish this here, it's better.

I believe that we have a different view on this. So yes, lets just leave it at that.

I might add though, that if the poster decides to buy just another ram stick, get one with the same timings at least.
 
I believe that we have a different view on this. So yes, lets just leave it at that.

I might add though, that if the poster decides to buy just another ram stick, get one with the same timings at least.
You are making a bigger deal out of memory compatibility than it actually is. If you are replacing or adding a DDR4 2400mhz module, nearly any of the same spec from any other brand will more than likely work fine and may even support tighter timings. It's a larger issue for system builders to use matching memory, but most of them have switched to single modules because speeds are much better now with DDR4 and it's cheaper to install a single module. This is also why people constantly reach out on forums for help with their slow computer when it turns out they have a single module. Generally the modules used by system builders are jedec standard modules. So nearly any DDR4 2133-2400mhz module will work with any other assuming they are of the same basic specs.
 
Last edited: