• Welcome to TechPowerUp Forums, Guest! Please check out our forum guidelines for info related to our community.
  • The forums have been upgraded with support for dark mode. By default it will follow the setting on your system/browser. You may override it by scrolling to the end of the page and clicking the gears icon.

MSI GeForce GTX 1650 Super Gaming X

W1zzard

Administrator
Staff member
Joined
May 14, 2004
Messages
28,831 (3.74/day)
Processor Ryzen 7 5700X
Memory 48 GB
Video Card(s) RTX 4080
Storage 2x HDD RAID 1, 3x M.2 NVMe
Display(s) 30" 2560x1600 + 19" 1280x1024
Software Windows 10 64-bit
NVIDIA's new GTX 1650 Super is 40% faster than the GTX 1650, which is a huge improvement. The card is priced competitively, too, obsoleting AMD's Radeon RX 580/590 offerings. MSI's Gaming X variant comes with a large cooler that offers great temperatures, extremely low gaming noise, and idle-fan-stop.

Show full review
 
Many reviews coming out today, stay tuned :)
 
personally i think AMD is hoping for nvidia to price their GTX1650S a bit higher. if not they might just as well release the card last month with the price tag of $150 and be done with it.
 
nice performance..

even in 4K, much better from 8GB cards ..

so the conclusion this is the best card for buck rightnow ?
 
Pitty for those 1660 user. Outdated so soon by a $60 cheaper GPU.
 
Curious to see how AMD will respond to this with RX 5500, cheaper? lower power draw? lets see they so much hyped 7nm, can they deliver? they have not far...
 
Curious to see how AMD will respond to this with RX 5500, cheaper? lower power draw? lets see they so much hyped 7nm, can they deliver? they have not far...
If you read this review, the 5500 is already in there.... power use on the AMD card is a bit more while performance is a bit lower.
 
If you read this review, the 5500 is already in there.... power use on the AMD card is a bit more while performance is a bit lower.
WoW! thanks, didnt know it was out yet! I would not say around 30% more power consumption and 20% less performance a bit lower though.
 
Its definitely concerning, from a technology point of view. The RX 5500 displays performance per watt of some "old" Pascal cards.

:shadedshu:
 
mmmmm I like That Average page :) Thx :)
 
So can we assume that all new cards even in the budget range will have GDDR6 memory?
 
WoW! thanks, didnt know it was out yet! I would not say around 30% more power consumption and 20% less performance a bit lower though.

What review did you read? I saw 13% more consumption and almost the same performance (a bit worse in some games).

1574437943637.png


1574438590777.png


It seems that the RX 5500 is an crap OEM model with reduced clocks. @W1zzard How can you come to a conclusion in the review without testing all games of the benchmark routine ?
 
I always check Furmark for Power Consumption it might not hold true for most of the games, but it shows what to expect, 103W for this vs 134 for rx 5500. that is almost 30% power consumption.
Not saying this is bad or good, in fact I really like the rx 5700, non xt versions. but it shows AMD is couple generations behind even with the new 7nm process at least on the GPU side.
 
My take aways ....

1. Since the 2xx series, AMD has very aggressively overclocked their cards "in the box". With the Super, it seems, nVidia is echoing this tactic.

2. I wouldn't expect RTX and DLSS to be available at this price point:

3. Thank You for this !

Video memory size of 4 GB might sound low at first, but you have to consider that pricing matters a lot in this segment. Adding more memory would make the card more expensive, with little or no performance difference at 1080p Full HD. Looking at our performance numbers, we can definitely see reduced FPS at 4K resolution compared to cards with more memory, but I'm not seeing anything in our data that would suggest these cards are memory-bound at 1080p.

It seems low only if one doesn't read TPU and other sitest test results. TPU testing for the 3GB and 6GB 1060s showed no discernable difference in performance between the 3GB and 6 GB models at 1080p / 1440p ... same for the 680, 770 and 960 2GB and 4 GB models ..... test results show that again her for the 1 650S.

At this point in time, I am getting concerned about nVidia's dominance across too many price niche's ... OK, nVidia grabbed ownership of the top tier when both the 780 and 780 Ti outperformed AMDs offerings ... but the, with 9xx, the 970 was a market monster, alone selling more than twice all AMD cards combined. The with 10xx , the 1060 dominated giving them sole ownership of the top 4 tiers. The 1060 has the largest market share of any card with 14.95% (580 has 1.6%). It's also the fastest growing in market share up 0.46% this month. However never did I think I'd see the day that nVidia would own the sub $200 market segment. Here's a comparison between the MSI1650 Super against what we can expect from asn MSI 5500 based u rations between MSI 5700

Performance ...

MSI 1650 Super = 100% from Performance Summary x (71.1 / 63.6) OC .. = 111.44
Tested 5500 = 94% x (119.6 / 115.1)* =97.68

MSI 1650 Super has estimated 14% performance advantage overclocked.

* OC results taken from TPUs MSI 5700 test (3.9% is typical of AMD cards. 5700xXT was 3.3%

AIB 5550 is expected to use 31.5% more power than MSI 1650 Super.

Power ...

MSI 1650 Super Peak Gaming is 105 watts
Reference 550 is 121 watts x (259 / 227)* = 138 watts

* Power increase for AIB card taken from comparison on reference 5700 and MSI 5700

Temperature ...

MSI 1650 Super hits 63C @ Full Load w/ OC
5500 Temps are unknown

Noise ...

The MSI 1650 Super hist 0 dbA at idle and 27C under full load
The 5500 hits 27 dbA at idle and 32C under full load

That makes the 5500 reference 41% louder

The MSI 5700 had the same sound level as the reference 5700 card

Obviously, the projected results above can not be confirmed until specific model testing is performed. But projections by W1zzard and the above are by no means "a guess" when substantial data sets are available. This is how science and technical evaluations work. You base your projections and similar models and when said models perform consistently across various data sets, you can be reasonably confident in your results.
 
Its definitely concerning, from a technology point of view. The RX 5500 displays performance per watt of some "old" Pascal cards.
The easiest explanation is that AMD sent the shittiest possible silicon to OEMs. I'm guessing the best parts went to Apple, and possibly into other laptops.
 
I always check Furmark for Power Consumption it might not hold true for most of the games, but it shows what to expect, 103W for this vs 134 for rx 5500. that is almost 30% power consumption.
Not saying this is bad or good, in fact I really like the rx 5700, non xt versions. but it shows AMD is couple generations behind even with the new 7nm process at least on the GPU side.

I do the same ... Furmark is a useful tool but the correct tool should be used for each job. Furmark is a great tool cooling system evaluation because it provides a consistent loading. You can evaluate various data points f the load is changing. ... setting up fan curves and tweaking you water pump speeds. But like Prime 95 and other synthetics .... it's a poor tool for testong OC stability or anything relative to actual usage. If you are trying to evaluate whether spending an extra $15 buying a Platinum model PSU over a Gold ... shouldm you base that decision on ....

a) A representative load which the PC will actually see pretty much every day for the next 4 years .... it
b) A synthetic load which the PC will never see after the 1st day or 2.

Yes, Furmark is a valuable tool ... but not for comparing real life power usage.
 
This is something GTX 1650 should've been already. Anyway, good review like always :)
 
Back
Top