');$('.tpu-fancybox-wrap').css('maxWidth', maxWidth);*/instance.$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, April 29th 2024
Exclusiveby btarunrDiscuss (57 Comments)
A few batches of GeForce RTX 4070 graphics cards are based on the 5 nm "AD103" silicon, a significantly larger chip than the "AD104" that powers the original RTX 4070. A reader has reached out to us with a curiously named MSI RTX 4070 Ventus 3X E 12 GB OC graphics card, saying that TechPowerUp GPU-Z wasn't able to detect it correctly. When we took a closer look at their GPU-Z submission data, we found that the card was based on the larger "AD103" silicon, looking at its device ID. Interestingly, current NVIDIA drivers, such as the 552.22 WHQL used here, are able to seamlessly present the card to the user as an RTX 4070. We dug through older versions of GeForce drivers, and found that the oldest driver to support this card is 551.86, which NVIDIA released in early-March 2024.
The original GeForce RTX 4070 was created by NVIDIA by enabling 46 out of 60 streaming multiprocessors (SM), or a little over 76% of the available shaders. To create an RTX 4070 out of an "AD103," NVIDIA would have to enable 46 out of 80, or just 57% of the available shaders, and just 36 MB out of the 64 MB available on-die L2 cache. The company would also have to narrow the memory bus down to 192-bit from the available 256-bit, to drive the 12 GB of memory. The PCB footprint, pin-map, and package size of both the "AD103" and "AD104" are similar, so board partners are able to seamlessly integrate the chip with their existing AD104-based RTX 4070 board designs. End-users would probably not even notice the change until they fire up diagnostic utilities and find them surprised.
Why NVIDIA would make RTX 4070 using the significantly larger "AD103" silicon, is anyone's guess—the company probably has a stash of chips that are good enough to match the specs of the RTX 4070, so it would make sense to harvest the RTX 4070 out of them, which could sell for at least $500 in the market. This also opens up the possibility of RTX 4070 SUPER cards based on this chip, all NVIDIA has to do is dial up the SM count to 56, and increase the L2 cache available to 48 MB. How the switch to AD103 affects power and thermals, is an interesting thing to look out for.
Our next update of TechPowerUp GPU-Z will be able to correctly detect RTX 4070 cards based on AD103 chips.
Related News
- Tags:
- AD103
- AD104
- GPU-Z
- MSI
- RTX 4070
- TechPowerUp
- Ventus
- WHQL
- Aug 21st 2023 NVIDIA BIOS Signature Lock Broken, vBIOS Modding and Crossflash Enabled by Groundbreaking New Tools (209)
- Dec 24th 2023 NVIDIA GeForce RTX 50 Series "Blackwell" On Course for Q4-2024 (126)
- May 5th 2024 NVIDIA to Only Launch the Flagship GeForce RTX 5090 in 2024, Rest of the Series in 2025 (152)
- Oct 17th 2023 NVIDIA Readies GeForce RTX 4070 SUPER, RTX 4070 Ti SUPER, and RTX 4080 SUPER (76)
- Aug 22nd 2023 NVIDIA Announces DLSS 3.5 Ray Reconstruction Technology, Works on GeForce 20 and Newer (89)
- Feb 19th 2024 NVIDIA RTX 50-series "Blackwell" to Debut 16-pin PCIe Gen 6 Power Connector Standard (106)
- Apr 5th 2024 NVIDIA Releases DLSS 3.7.0 With Quality E Preset for Image Quality Improvements (54)
- Jun 26th 2023 More Pictures of NVIDIA's Cinder Block-sized RTX 4090 Ti Cooler Surface (145)
- Apr 12th 2024 NVIDIA Points Intel Raptor Lake CPU Users to Get Help from Intel Amid System Instability Issues (106)
- Dec 14th 2023 NVIDIA GeForce RTX 4080 SUPER, RTX 4070 Ti SUPER, and RTX 4070 SUPER Release Dates Leaked (49)
mubarak ahmedI tried searching the box for any manufacturing date but I didn't find anything, and the BIOS is not available in the database
Please upload that BIOS to help W1zz in adding support for this card to GPU-Z.
stimpy88And the performance of the non-RT hardware in the newer cards is not having an impact... :kookoo: Clockspeeds, shaders, VRAM bandwidth etc? Try looking at the performance hit from enabling RT.
You were complaining solely about RT performance stagnation, so I showed that the RT performance jumping an enormous 83% and 76% amount per generation.
RT performance is improving faster than non-RT performance, since general performance per gen including all the raster/shader/FLOPS stuff is more like 25-50% depending on which models you pick to compare.
So even relative to the non-RT performance gains per generation, the RT hardware is improving at about double that rate. How is that a "glacial pace" of RT improvement?
Feel free to explain it in a way that makes sense, but as far as I can see it's pretty black and white; RT performance is improving faster than anything else in the GPU space right now. It's the opposite of glacial.
AssimilatorPlease upload that BIOS to help W1zz in adding support for this card to GPU-Z.
I've already uploaded it to the database, but just to make sure nothing I made was wrong I will upload it here because I saw the PCIe is set to gen 3 and the card it self is gen 3 also.
mubarak ahmedI've already uploaded it to the database, but just to make sure nothing I made was wrong I will upload it here because I saw the PCIe is set to gen 3 and the card it self is gen 3 also.
That's due to your motherboard/CPU, one or both of which doesn't support PCIe 4.0.
AssimilatorThat's due to your motherboard/CPU, one or both of which doesn't support PCIe 4.0.
I know but it should say that the card supports PCIE 4.0 but it's running on 3.0 right?
mubarak ahmedI know but it should say that the card supports PCIE 4.0 but it's running on 3.0 right?
Yes, you are right! Maybe the fact that GPU-Z doesn't support this model is affecting its ability to detect the bus speed correctly.
mubarak ahmedI know but it should say that the card supports PCIE 4.0 but it's running on 3.0 right?
That depends. What CPU/motherboard do you have?
