');$('.tpu-fancybox-wrap').css('maxWidth', maxWidth);*/instance.$refs.stage.on('transitionend', function() {updateButtonPos(instance);});},onUpdate: updateButtonPos,afterShow: function(instance, slide) {updateButtonPos(instance);instance.$refs.inner.find('.fancybox-tpu-nav').show();},beforeClose: function(instance, slide) {instance.$refs.inner.find('.fancybox-tpu-nav').hide();},afterClose: function(instance, slide) {$('.tpu-fancybox-wrap').contents().unwrap();$('body').removeClass('tpu-fancybox-body-wrap')},baseTpl: '
' +'
' +'
' +'
' +'' +'' +'' +'
' +'
' +'
' +'
' +'
',});});}loadjs.ready(['jquery', 'fancybox', 'swiper'], function() {attachLightbox('a[data-fancybox]');if ($(window).width()<600) {$('.imgcontainer').each(function() {var $this=$(this);if (($this.find('a').length==1) || ($this.find('a').length>7))return;$this.addClass('swiper-container');$this.find('a').addClass('swiper-slide').css('width', 'auto').wrapAll('
');$this.find('.swiper-wrapper').after('
');new Swiper ($this.eq(0), { slidesPerView: 'auto', slidesPerGroup: 1, spaceBetween: 15, pagination: { el: '.swiper-pagination', clickable: true } });});}$('.newspost').on('click', '.spoiler > .button, .spoiler > a', function(e) {e.preventDefault();$(this).next('div').toggle();});$('.newspost').on('click', '.ispoiler', function(e) {e.preventDefault();$(this).find('div').css('filter', '');$(this).removeClass('ispoiler');});$('.contnt').on('click', '.newspoll_btn', function() {popup.Show('TechPowerUp Quick Poll','Loading...');$.get('/news-poll/options?id='+$(this).data('id'), function(data) {$('#popup_content').html(data);});});});
Monday, February 7th 2022
by UskompufDiscuss (18 Comments)
The recently announced entry-level NVIDIA GeForce MX550 Turing-based discrete mobile graphics card for thin and light laptops has recently appeared on the PassMark video card benchmark site. The MX550 scores 5014 points in the G3D Mark test which places its performance nearly exactly with that of the integrated Vega 8 iGPU found in the Ryzen 9 5900HS that scores 4968 points in the same benchmark. There is only a single test result available for the MX550 so we will need to wait for further benchmarks to confirm its exact performance but either way it represents a significant performance improvement from the MX450 which scores just 3724 points. The MX550 is a PCIe 4.0 card featuring the 12 nm TU117 Turing GPU with 1024 shading units paired with 2 GB of GDDR6 memory.
Related News
- Tags:
- MX550
- NVIDIA
- PassMark
- Turing
- Aug 21st 2023 NVIDIA BIOS Signature Lock Broken, vBIOS Modding and Crossflash Enabled by Groundbreaking New Tools (210)
- Dec 24th 2023 NVIDIA GeForce RTX 50 Series "Blackwell" On Course for Q4-2024 (126)
- May 24th 2024 NVIDIA RTX 5090 "Blackwell" Founders Edition to Implement the "RTX 4090 Ti" Cinderblock Design (118)
- May 5th 2024 NVIDIA to Only Launch the Flagship GeForce RTX 5090 in 2024, Rest of the Series in 2025 (154)
- Jul 15th 2024 NVIDIA GeForce RTX 50 Series "Blackwell" TDPs Leaked, All Powered by 16-Pin Connector (167)
- Feb 19th 2024 NVIDIA RTX 50-series "Blackwell" to Debut 16-pin PCIe Gen 6 Power Connector Standard (106)
- Aug 22nd 2023 NVIDIA Announces DLSS 3.5 Ray Reconstruction Technology, Works on GeForce 20 and Newer (89)
- Apr 5th 2024 NVIDIA Releases DLSS 3.7.0 With Quality E Preset for Image Quality Improvements (54)
- Jun 11th 2024 Possible Specs of NVIDIA GeForce "Blackwell" GPU Lineup Leaked (141)
- Oct 17th 2023 NVIDIA Readies GeForce RTX 4070 SUPER, RTX 4070 Ti SUPER, and RTX 4080 SUPER (76)
wait that cant be right, the mx550 looks like a rebranded 1650m
that should be like, significantly more powerful than the anemic vega 8
Meaning this is what nvidia wants to charge sheeps, mx550 for 600 usd hehe
Freshwater Moderator
Selayawait that cant be right, the mx550 looks like a rebranded 1650m
that should be like, significantly more powerful than the anemic vega 8
I gave up on making sense of laptop parts, because you can get the same exact GPU model but mystery TDP values and RAM width that gives you 5 different performance variants of the one model
Vega 8 is a integrated GPU. Why would you compare a desktop or mobile part for that matter? Vega is around for some being as well. The RDNA(2) is going to take over in APU's.
The headline should be "loses to iGPU". If this is true performance then it's very slow for a dGPU. There is also the MX570 based on GA108 that should fare better since MX550 is based on old Turing TU117.
I will take the results from benchmark with a big pinch of salt. In my past testing comparing a Vega 11 (Ryzen 5 3400G) with the MX450, the former is almost as quick when running benchmarks, i.e. 3D Mark and Unigine Heaven. But in games, the dedicated GPU tends to pull ahead, unless it is let down by poor cooling. The Nvidia MX series GPUs are always used in slim laptops, thus, the cooling solution is generally the reason for poor performance due to throttling. Other reason is when one is gaming at a setting that is too high for the GPU due to the 2GB dedicated VRAM. As a result, performance will plummet once it exceeds the available dedicated VRAM. In 2022, I feel it makes no sense to keep 2GB VRAM for the MX series since it is quite common for newer games to be quite VRAM hungry.
Selayawait that cant be right, the mx550 looks like a rebranded 1650m
You can thank Nvidia for that, they made sure to destroy any kind of connection between nomenclature and expected performance.
So a video card with a cut down GPU that is relatively easy to produce matches the performance (in a synthetic benchmark) of an integrated GPU that was released a year ago.
In games that the 2GB won't kill the performance (Vega supports 4GB) it should be on average at least +20% if running at base always, up to +45% when running in turbo always.
TPU places it at RX570 -20% performance level (1080p) which I think is optimistic:
www.techpowerup.com/gpu-specs/geforce-mx550.c3860
Shouldn't the title be the other way around? Like "iGPU in vega renders semi-dedicated GPU named MX550 obsolete"?
As per Dr Su, 19 million of notebooks with MX bazinga are sold around annually (or was it quarterly? yikes)
Selayawait that cant be right, the mx550 looks like a rebranded 1650m
that should be like, significantly more powerful than the anemic vega 8
The MX550 is not close to the 1650m:
Half the board power (so lower clocks)
Half the TMUs
Half the ROPs
Half the memory
Half the memory bandwidth
I know, still should like blow up the vega 8 without breaking a sweat
i mean, the mx450 is like, totally trouncing the vega 8 already so yeah
I dont know what to think of these results, does it complement how good the 5900HS GPU is or is it saying the MX550 is just that bad.
Freshwater Moderator
GhostRyderI dont know what to think of these results, does it complement how good the 5900HS GPU is or is it saying the MX550 is just that bad.
Pro nvidia website wants to make AMD look bad, confuses everyone instead
I've seen this 'news' reposted to a dozen places, and all the comments are asking "what kind of f*cked up comparison is this?"
JismVega 8 is a integrated GPU. Why would you compare a desktop or mobile part for that matter? Vega is around for some being as well. The RDNA(2) is going to take over in APU's.
Think it's just for comparison. If i buy this it'll perform much the same as this. I wouldn't read too much into it.
MusselsPro nvidia website wants to make AMD look bad, confuses everyone insteadI've seen this 'news' reposted to a dozen places, and all the comments are asking "what kind of f*cked up comparison is this?"
To me it just says 'Buy the AMD APU' over this.
MusselsPro nvidia website wants to make AMD look bad, confuses everyone insteadI've seen this 'news' reposted to a dozen places, and all the comments are asking "what kind of f*cked up comparison is this?"
Actually it makes the new MX 550 looks bad, because it barely beats an integrated GPU. Mind you, without contest it's hard to tell that, for instance, the Vega 7 of a more medium range 5600U gets just 2101 on Passmark. But honestly you know what I find even more puzzling than this piece of news appearing everywhere? A staff member criticising it after the TPU staff just reposted it.
So nvidia made a dGPU that matches an iGPU in speed. Um... wow... I guess? :wtf:
