This review tests the MSI GeForce RTX 3080 Ti SUPRIM X; yes, this is a factory tweaked 3080 Ti that brings in an additional 5% extra performance whilst merged with a custom PCB and cooling design. It's a graphics card series that stays close to 3090 performance levels. Yeah, there's no denying it. It has been a bizarre year if we're talking computer components alone, silicon shortages, mining; it's been sour grapes for a long time now. Likely that won't even change soon. People are almost fighting to be able to grab a decently proceed graphics card, often running up to three times the MSRP pricing. NVIDIA is staying on its trajectory, though, implementing hash-rate limitations for consumer-grade GPUs and releasing refresh products pretty much as scheduled. Today, we see such a refresh product that positioned itself between the GeForce RTX 3080 and 3090. That product would be the GeForce RTX 3080 ti, and probably was the worst kept secret as of late in the history of NVIDIA. Armed with a shader core count of 10240 units, this card is paired with 12 GB of GDDR6X graphics memory running 19 Gbps. Much has been said, rumored, and spoken about this card, as with that libidinous shading core count, it is bound to be a bit of a beast in the enthusiast performance segment. You need to remember that it was less than a year ago that NVIDIA launched their Ampere GPU series, September 2020, in fact. A week before announcements, specifications of the GeForce RTX 3060 Ti, 3070, 3080, and 3090 took a twist; the shader core count mysteriously doubled up from what everybody expected, to date something massively important to this product range as yes, the competition has gotten fiercer as well. NVIDIA's GPUs are fabricated on an 8nm node derived from Samsung. This process further develops Samsung's 10nm process; no EUV is applied in production just yet. This second wave of announcements will see the GeForce RTX 3080 Ti and next week 3070 Ti (hey, come on, it's not a secret). The desktop product line of Ampere for consumers now entails the GeForce RTX 3060 Ti 8GB, 3070 8GB GDDR6, RTX 3080 10GB, RTX 3080 Ti 12GB, and RTX 3090; what we test today is a 12GB GDDR6X based close to premium flagship. Much like the 3080 and 3090, the 3080 Ti will base based on the existing 28 billion transistors based GA102 GPU from NVIDIA, obviously reconfigured. The card has an impressive 10240 Shader processors activated, and mind you; the 3090 has 10496. A big change is that the memory is halved from 24 towards 12 GB (which is an excellent number). RTX 3080 Ti SUPRIM X

Don't sweat it, Airflow Control guides the air to exactly where it needs to be for maximum cooling. Phần tản nhiệt được thiết kế chuẩn xác để tiếp xúc tối đa với GPU giúp truyền nhiệt một cách tối ưu nhất. Luin useamman kerran ja vaikka sieltä löytyy mm. tämä "Ylikellotukseen 3080 Suprim X soveltuu korkean 430 watin Power Limit -tehonkulutusrajoituksen myötä hyvin ja on samassa kategoriassa Asuksen ROG Strix -mallin kanssa. Grafiikkapiiri saatiin toimimaan vakaasti 2070-2085 MHz:n kellotaajuudella ", niin olisin nähnyt testin järkevämpänä samantapaisten korttien kesken. The second photo is a close-up shot of the middle portion of the graphics card. I have not dismantled the graphics card but you can see from that angle several heat pipes and the dedicated heatsink for the VRAM.

We are Sorry !

Among the three games above, I think Cyberpunk 2077 has the best implementation of ray tracing and DLSS. Cyberpunk 2077 features realistic reflections, improved lighting, enhanced shadowing, and refined ambient occlusion. As a result, the game is very graphics intensive, especially if you enable all of these visual effects and set to ultra. When it comes to performance, well, you have seen the results above. The RTX 3080 SUPRIM X performs a bit faster compared to a reference-spec RTX 3080 like the GALAX SG. This is thanks to its better cooling solution and higher power limit. Although, honestly, the difference isn’t that huge or noticeable without benchmarking and comparing the numbers side by side.

Suprim X:n BIOS-kytkimen myötä vaihtoehtoinen Gaming-tila tiputtaa grafiikkapiirin lämpötila 69 asteeseen ja GDDR6X-muistien lämpötila 12 asteella 96 asteesta 86 asteeseen, mutta samalla melutaso nousee 3 desibeliä äänekkäämmäksi 39,5 desibeliin. Asuksen TUF OC:n Performance-tilaan verrattuna Suprim X:n Gaming-tilan grafiikkapiirin lämpötila on 6 astetta korkeampi, muistien 2 astetta viileämpi ja melutaso 1,5 desibeliä äänekkäämpi. A masterpiece of teamwork, fan blades work in pairs to create unprecedented levels of focused air pressure. For this test, I used the NVIDIA driver version 457.30. The graphics card was tested in three resolutions – 1920×1080 FHD, 2560×1440 WQHD, and 3840×2160 4K UHD. When the RTX 20 series launched, there were no games that fully supported ray tracing. And some of the early games that did support them wasn’t visually jaw-dropping, at least in my opinion. Also, the first version of DLSS was bad that I didn’t recommend using it at all. Ennen näytönohjaimen esittelyä ja testeihin siirtymistä on kuitenkin puhuttava saatavuudesta ja hinnasta. GeForce RTX 3080 Suprim X saapui Suomessa myyntiin marraskuun lopulla 949 euron hinnalla, jolloin se oli lähes yhtä kallis kuin Asuksen ROG Strix ja noin 50 euroa kalliimpi kuin Gaming X Trio ja Asuksen TUF OC.

Render Config

GDDR6X-muistien lämpötila oli 3080 Suprim X:llä vakiona eli Silent-tilassa 96 astetta eli neljä astetta alhaisempi kuin Gaming X Triolla, jolla muistien lämpötila nousi 100 asteeseen. Gaming-tilassa Suprim X:llä muistien lämpötila laski 10 asteella 86 asteeseen eli kaksi astetta alhaisemmaksi kuin Asuksen TUF OC:lla Performance-tilassa. With the GeForce RTX 3080 SUPRIM X 10G graphics card, I think MSI has “supercharged” NVIDIA’s RTX 3080 GPU. It was built with superb craftsmanship, exceptional aesthetics with meticulous attention to detail. Tosin itselläni nuo siilit ja propelit ovat luultavasti turhia takalevyä lukuunottamatta, sillä joku 3080 riisutaan ja menee joka tapauksessa vedelle. Grafiikkapiiri ylikellottui +125 MHz eli 2030 MHz:n Boost-taajuudelle, jolloin todellinen kellotaajuus Battlefield V:ssä 4k-resoluutiolla vaihteli noin 2070-2085 MHz:n välimaastossa. GDDR6X-muistit venyivät vakaasti ilman ongelmia +750 MHz eli 20,5 Gbps:n nopeudelle.