GeForce GTX 1660 Tested: 33 Game Benchmark feat. The Division 2

[Julio Franco]

This is our second look at the new GeForce GTX 1660. Not to be confused with the 1660 Ti that was released a month earlier, both GPUs offer great value at mid-range prices of $220 for the GTX 1660 and $280 for the Ti version.

Read the full article here.

https://www.techspot.com/review/1813-geforce-gtx-1660-mega-benchmark/

 

[krizby]

Buy the cheapest 1660 you can find, flash it with EVGA 1660 XC Ultra 160W TDP bios for an easy 15% bump in performance lol. Flashing the quite simple and literally takes like 5 mins.

 

[Puiu]

I was really expecting a bigger difference, but it is still a good buy, unless AMD manages to impress at this price point with Navi this summer.

 

[mrtraver]

Sounds like a nice bump up from my GTX 480. The Forza Horizon 4 demo won't even open on my machine.

 

[birthdaymonkey]

Buy the cheapest 1660 you can find, flash it with EVGA 1660 XC Ultra 160W TDP bios for an easy 15% bump in performance lol. Flashing the quite simple and literally takes like 5 mins.

I just ordered that GPU. Is this 160w TDP BIOS the stock one that comes with the card?

According to Techpowerup, the EVGA 1660 Ultra performs about the same as the MSI Gaming X. Was the Gaming X used for this review, or just for the photo at the top of the article?

An extra 15% (close to 1070/stock 1660Ti) would be nice.

I've also read that the 1660 is memory bandwidth starved, so I plan to bump the GDDR5 clock as high as possible.

 

[Cycloid Torus]

Since we are just now emerging from the crypto-currency follies, the key question for older cards is, "Are they all still overpriced?"

Current spread of cost per frame is from less that $2/fps to almost $5/fps. Should the high end be so expensive or is this just 'marketing'? For instance, why isn't the GTX 1060 6G selling for just under $200 rather than $240? Or, the GTX 1070 at $260 instead of $330? These are older cards with GDDR5 - not the new stuff.

 

[krizby]

I just ordered that GPU. Is this 160w TDP BIOS the stock one that comes with the card?

According to Techpowerup, the EVGA 1660 Ultra performs about the same as the MSI Gaming X. Was the Gaming X used for this review, or just for the photo at the top of the article?

An extra 15% (close to 1070/stock 1660Ti) would be nice.

I've also read that the 1660 is memory bandwidth starved, so I plan to bump the GDDR5 clock as high as possible.

Yes at stock (100% Power Limit) the Msi Gaming X and EVGA XC Ultra should be identical in performance since they are both 130W TDP, however the EVGA XC Ultra allows power limits to be increased to 160W (126% PL) while the Gaming X allows only 140W (108% PL).

Well I think the best model is the MSI 1660 Ventus at 229usd or the Gigabyte version at 219$, flash it with the EVGA XC Ultra bios that is available on Techpowerup and you pretty much save yourself 20/30$. Use the 30$ you save and buy some Thermal Grizzly Conductonaut for better cooling results and some fun shunt modding the 1660.

 

[bluetooth fairy]

It was pretty shocking to see the GTX 1660’s lead over the GTX 1060 6GB shrivel up from 21% to just 14%.

Is there any sense to use median, std deviation for such of a data? I just thought there could be a measure in shock prediction. May be not for public use, but as an internal triple-checking.

 

[Kytetiger]

Is there a way to see the impact of GDDR5 vs GDDR6 only?
To see if the RAM is the major reason for this performance vs the TI.

 

[Vulcanproject]

Is there a way to see the impact of GDDR5 vs GDDR6 only?
To see if the RAM is the major reason for this performance vs the TI.

Well the core is slightly cut down, so that is one reason. About 8 percent less CUDA cores. Can be mitigated by increasing the core clocks.

Overclocking the hell out of the GDDR5 modules on the non Ti brings very nice gains. You can get very close performance to a stock GTX1660Ti. It's obviously a major factor.

 

[DiGiTAL80]

Why is the vega 56 and 64 power consumption almost the same?

 

[Nightfire]

Why is the vega 56 and 64 power consumption almost the same?

I know on older games, like Crysis 3 which was used, the Vega 64 had poor optimization over Vega 56 where it often yielded similar performance.

 

[Nightfire]

Steve, maybe you should double check Division 2. Your graph shows the 1060 6GB and 3GB being virtually the same. I have never seen that in a game since the last 6 GB card has more Cuda cores.

 

[krizby]

Why is the vega 56 and 64 power consumption almost the same?

Probably a typo lol, Vega 56 should pull 100W less than Vega 64. The man really needs some sleep .

 

[Steve]

Why is the vega 56 and 64 power consumption almost the same?

Probably a typo lol, Vega 56 should pull 100W less than Vega 64. The man really needs some sleep .

Nope, the ASUS Strix models are much the same.

 

[Do Mor3 G]

I was really expecting a bigger difference, but it is still a good buy, unless AMD manages to impress at this price point with Navi this summer.

Indeed, I recon if the "GTX 1080" performance rumors do hold true, it will be at least better than anything the current Polaris range has to offer, basically providing Vega 56 like performance in a possibly cheaper bracket. The GTX 16X0 line will have serious competition if Navi scales accordingly...

But this is all hypothetical anyway till the Navi Fat Lady sings

 

[megamaan]

Buy the cheapest 1660 you can find, flash it with EVGA 1660 XC Ultra 160W TDP bios for an easy 15% bump in performance lol. Flashing the quite simple and literally takes like 5 mins.

Hello there!

Have you tried it? This is safe? I have a GTX 1660 Zotac AMP, and I would like to try the bios of the EVGA card, but I am afraid that my card will stop working and die.

Thanks.