KitGuru KitGuru.net – Tech News | Hardware News | Hardware Reviews | IOS | Mobile | Gaming | Graphics Cards
Our acoustic measurements are less precise on this mid-range test system, the noise floor of the totally quiet testing room is 35 dBA as measured with a Benetech GM1351 Digital Sound Level Meter.
We take our measurements with the decibel meter protruding off the front-top section of the chassis, overhanging by exactly 1 inch. The underlying noise level of the system, generated by the 800 RPM Noctua case fans and H100i with fixed 700RPM fans is 39 dBA, thus anything above this level can be attributed to the graphics card since the PSU is passive for most of it's power capability.
Noise levels were measured after 10 minutes of load under three scenarios: Furmark GPU stress test, Unigine Valley looping at the Extreme HD preset and desktop idle.
There is no apparent change in system noise level irrespective of the 3D load placed upon the GTX 950-2G, the combination of a low TDP and dual-fan cooler keeps things cool and quiet.
For reference the fan reached 39% (1,350 RPM) under the most stressful load scenario – Furmark. Considering these are small fans, 70mm, the noise output is lower than an equivalent RPM on a larger fan.
Curious timing for this review with the low power RX cards due soon. It would be good to see a follow up review comparing this to those.
Unfortunately people buying this card will be building on a budget and won’t be running watercooled intel i7’s on an SSD with 32gb of high speed DDR4 ram, meaning the results will be VERY different for them.
I know it has been reviewed this way to show the cards true potential without any bottlenecking, but if anybody sees these results, then buys the gpu to pair with their £60 stock cooled amd processor and 4gb of RAM, running on a 7200rpm HDD disk then they are going to be devastated when they get massively worse results.
Personally, i’d rather see GPU’s tested in the average setup they will be expected to be bought for.
Thanks for the feedback, it is a tricky situation – keeping a test bench consistent between a wide range of products (as low as R7 360 but as high as R9 390) but also needing to make sure it matches the target audience for a given product, such as in this case. Having test benches that reflect different price points (low-mid range, mid to high-end) is something I’ll (re)consider going forward as I have done this before in my reviewing career.
Definitely, we’ll try and take a look at as many new products as possible including anything new from AMD in the same power/performance segment.
Definitely something i’d look forward to seeing. I would also find it very interesting to see how much of a bottleneck a budget setup vs a test rig like this would be, but that also doubles the amount of work required for a review.
Either way, it is a very good review, and I like seeing the lower end products getting highlighted.
Thanks for the response Ryan.
I am earning around 6000-8000 dollars a month for freelancing i do from my home. For everybody prepared to do simple online jobs for few hours a day from your couch at home and make solid payment while doing it… Try this work SELF90.COM
I’m running an Elsa GTX 950 2GB (Great Japanese brand, no power connector) on a 10 year old Asus P5 Mobo Win 10 X64 Pro with a Q9650 and 8gb cheapo ram (Nanya PC2-6400 DDR2-SDRAM) and it runs Doom 2016 Vulkan on medium at ~45fps. No Mans Sky, XCOM 2 also perform really well on medium settings. Bear in mind it’s running in a PCI Express 1.1 slot too.
I had an Asus GTX 750Ti in there but it started going crazy with the fan going on and off, and I found this Elsa 950 in Akihabara going really cheap. Knocks the 750Ti out of the park.
I think it’s this one: http://www.elsa-jp.co.jp/products/products-top/graphicsboard/geforce/midrange/geforce_gtx950_2gb/