Podcast #514 - NVIDIA GeForce RTX 2080 and 2080 Ti Deep Dive

Subject: General Tech | September 20, 2018 - 12:23 PM |
Tagged: turing, RTX 2080 Ti, RTX 2080, rtx, podcast, pascal, nvidia, Intel, i9-9900K, i7-9700K, coffee lake

PC Perspective Podcast #514 - 09/20/18

Join us this week for discussion on both the Turing architecture, NVIDIA RTX 2080 and 2080 Ti product reviews, more 8-core Intel Coffee Lake Rumors and more!

You can subscribe to us through iTunes and you can still access it directly through the RSS page HERE.

The URL for the podcast is: http://pcper.com/podcast - Share with your friends!

Hosts: Ryan Shrout, Jeremy Hellstrom, Josh Walrath, Allyn Malventano

Peanut Gallery: Alex Lustenberg

Program length: 1:38:19

Podcast topics of discussion:

  1. Week in Review:
  2. Thanks to Casper for supporting our podcast! Save $50 on select mattresses at http://www.casper.com/pcper code pcper
  3. News items of interest:
  4. Picks of the Week:
    1. 1:26:40 Jeremy: @#*$ing NewEgg
    2. 1:28:35 Josh: Physical Media 4 Evah
    3. 1:31:20 Allyn: Lego Voltron
  5. Closing/outro

Author:
Manufacturer: NVIDIA

A Look Back and Forward

Although NVIDIA's new GPU architecture, revealed previously as Turing, has been speculated about for what seems like an eternity at this point, we finally have our first look at exactly what NVIDIA is positioning as the future of gaming.

geforce-rtx-2080.png

Unfortunately, we can't talk about this card just yet, but we can talk about what powers it

First though, let's take a look at the journey to get here over the past 30 months or so.

Unveiled in early 2016, Pascal marked by the launch of the GTX 1070 and 1080 was NVIDIA's long-awaited 16nm successor to Maxwell. Constrained by the oft-delayed 16nm process node, Pascal refined the shader unit design original found in Maxwell, while lowering power consumption and increasing performance.

Next, in May 2017 came Volta, the next (and last) GPU architecture outlined in NVIDIA's public roadmaps since 2013. However, instead of the traditional launch with a new GeForce gaming card, Volta saw a different approach.

Click here to continue reading our analysis of NVIDIA's Turing Graphics Architecture

NVIDIA Launches GTX 1050 3GB for Budget Gamers

Subject: Graphics Cards | May 23, 2018 - 06:21 PM |
Tagged: pascal, nvidia, GP107, GDDR5, budget

NVIDIA recently quietly launched a new budget graphics card that neatly slots itself between the GTX 1050 and the GTX 1050 Ti. The new GTX 1050 3GB, as the name suggests, features 3GB of GDDR5 memory. The new card is closer to the GTX 1050 Ti than the name would suggest, however as it uses the same 768 CUDA cores instead of the 640 of the GTX 1050 2GB. The GDDR5 memory is where the card differs from the GTX 1050 Ti though as NVIDIA has cut the number of memory controllers by one along with the corresponding ROPs and cache meaning that the new GTX 1050 3GB has a smaller memory bus and less memory bandwidth than both the GTX 1050 2GB and GTX 1050 Ti 4GB.

NVIDIA GTX 1050.png

Specifically, the GTX 1050 with 3GB GDDR5 has a 96-bit memory bus that when paired with 7 Gbps GDDR5 results in maximum memory bandwidth of 84 GB/s versus the other previously released cards' 128-bit memory buses and 112 GB/s of bandwidth.

Clockspeeds on the new GTX 1050 3GB start are a good bit higher than the other cards though with the base clocks starting at 1392 MHz which is the boost clock of the 1050 Ti and running up to 1518 MHz boost clockspeeds. Thanks to the clockspeeds bumps, the theoretical GPU performance of 2.33 TFLOPS is actually higher than the GTX 1050 Ti (2.14 TFLOPS) and existing GTX 1050 2GB (1.86 TFLOPS) though the reduced memory bus (and loss of a small amount of ROPs and cache) will hold the card back from surpassing the Ti variant in most workloads – NVIDIA needs to maintain product segmentation somehow!

  NVIDIA GTX 1050 2GB NVIDIA GTX 1050 3GB NVIDIA GTX 1050 Ti 4GB AMD RX 560 4GB
GPU GP107 GP107 GP107 Polaris 11
GPU Cores 640 768 768 896 or 1024
Texture Units 40 48 48 64
ROPs 32 ? 32 16
GPU Base 1354 1392 1290 1175
GPU Boost 1455 1518 1392 1275
TFLOPS 1.86 2.33 2.14 up to 2.6
Memory 2GB GDDR5 3GB GDDR5 4GB GDDR5 2GB or 4GB GDDR5
Memory Clockspeed 7 Gbps 7 Gbps 7 Gbps 7 Gbps
Memory Bus 128-bit 96-bit 128-bit 128-bit
Memory Bandwidth 112 GB/s 84 GB/s 112 GB/s 112 GB/s
TDP 75W 75W 75W 60W to 80W
Pricing ~$150 ~$160 (Estimate) ~$200 ~$160

The chart above compares the specifications of the GTX 1050 3GB with the GTX 1050 and the GTX 1050 Ti on the NVIDIA side and the AMD RX 560 which appears to be its direct competitor based on pricing. The new 3GB GTX 1050 should compete well with AMD's Polaris 11 based GPU as well as NVIDIA's own cards in the budget gaming space where hopefully the downside of a reduced memory bus will at least dissuade cryptocurrency miners from adopting this card as an entry level miner for Ethereum and other alt coins giving gamers a chance to buy something a bit better than the GTX 1050 and RX 550 level at close to MSRP while the miners fight over the Ti and higher variants with more memory and compute units.

NVIDIA did not release formal pricing or release date information, but the cards are expected to launch in June and prices should be around $160 to $180 depending on retailer and extra things like fancier coolers and factory overclocks.

What are your thoughts on the GTX 1050 3GB? Is it the bastion of hope budget gamers have been waiting for? hehe Looking around online it seems pricing for these budget cards has somewhat returned to sane levels and hopefully alternative options like these aimed at gamers will help further stabilize the market for us DIYers that want to game more than mine. I do wish that NVIDIA could have changed the name a bit to better differentiate the card, maybe the GTX 1050G or something but oh well. I suppose so long as the 640 CUDA core GTX 1050 doesn't ever get 3GB GDDR5 at least gamers will be able to tell them apart by the amount of memory listed on the box or website.

Also read:

Source: NVIDIA
Author:
Manufacturer: NVIDIA

Looking Towards the Professionals

This is a multi-part story for the NVIDIA Titan V:

Earlier this week we dove into the new NVIDIA Titan V graphics card and looked at its performacne from a gaming perspective. Our conclusions were more or less what we expected - the card was on average ~20% faster than the Titan Xp and about ~80% faster than the GeForce GTX 1080. But with that $3000 price tag, the Titan V isn't going to win any enthusiasts over.

What the Titan V is meant for in reality is the compute space. Developers, coders, engineers, and professionals that use GPU hardware for research, for profit, or for both. In that case, $2999 for the Titan V is simply an investment that needs to show value in select workloads. And though $3000 is still a lot of money, keep in mind that the NVIDIA Quadro GP100, the most recent part with full-performance double precision compute from the Pascal chip, is still selling for well over $6000 today. 

IMG_5009.JPG

The Volta GV100 GPU offers 1:2 double precision performance, equating to 2560 FP64 cores. That is a HUGE leap over the GP102 GPU used on the Titan Xp that uses a 1:32 ratio, giving us just 120 FP64 cores equivalent.

  Titan V Titan Xp GTX 1080 Ti GTX 1080 GTX 1070 Ti GTX 1070 RX Vega 64 Liquid Vega Frontier Edition
GPU Cores 5120 3840 3584 2560 2432 1920 4096 4096
FP64 Cores 2560 120 112 80 76 60 256 256
Base Clock 1200 MHz 1480 MHz 1480 MHz 1607 MHz 1607 MHz 1506 MHz 1406 MHz 1382 MHz
Boost Clock 1455 MHz 1582 MHz 1582 MHz 1733 MHz 1683 MHz 1683 MHz 1677 MHz 1600 MHz
Texture Units 320 240 224 160 152 120 256 256
ROP Units 96 96 88 64 64 64 64 64
Memory 12GB 12GB 11GB 8GB 8GB 8GB 8GB 16GB
Memory Clock 1700 MHz MHz 11400 MHz 11000 MHz 10000 MHz 8000 MHz 8000 MHz 1890 MHz 1890 MHz
Memory Interface 3072-bit
HBM2
384-bit G5X 352-bit G5X 256-bit G5X 256-bit 256-bit 2048-bit HBM2 2048-bit HBM2
Memory Bandwidth 653 GB/s 547 GB/s 484 GB/s 320 GB/s 256 GB/s 256 GB/s 484 GB/s 484 GB/s
TDP 250 watts 250 watts 250 watts 180 watts 180 watts 150 watts 345 watts 300 watts
Peak Compute 12.2 (base) TFLOPS
14.9 (boost) TFLOPS
12.1 TFLOPS 11.3 TFLOPS 8.2 TFLOPS 7.8 TFLOPS 5.7 TFLOPS 13.7 TFLOPS 13.1 TFLOPS
Peak DP Compute 6.1 (base) TFLOPS
7.45 (boost) TFLOPS
0.37 TFLOPS 0.35 TFLOPS 0.25 TFLOPS 0.24 TFLOPS 0.17 TFLOPS 0.85 TFLOPS 0.81 TFLOPS
MSRP (current) $2999 $1299 $699 $499 $449 $399 $699 $999

The current AMD Radeon RX Vega 64, and the Vega Frontier Edition, all ship with a 1:16 FP64 ratio, giving us the equivalent of 256 DP cores per card.

Test Setup and Benchmarks

Our testing setup remains the same from our gaming tests, but obviously the software stack is quite different. 

  PC Perspective GPU Testbed
Processor Intel Core i7-5960X Haswell-E
Motherboard ASUS Rampage V Extreme X99
Memory G.Skill Ripjaws 16GB DDR4-3200
Storage OCZ Agility 4 256GB (OS)
Adata SP610 500GB (games)
Power Supply Corsair AX1500i 1500 watt
OS Windows 10 x64
Drivers AMD: 17.10.2
NVIDIA: 388.59

Applications in use include:

  • Luxmark 
  • Cinebench R15
  • VRay
  • Sisoft Sandra GPU Compute
  • SPECviewperf 12.1
  • FAHBench

Let's not drag this along - I know you are hungry for results! (Thanks to Ken for running most of these tests for us!!)

Continue reading part 2 of our Titan V review on compute performance!!

Gigabyte Launches GTX 1070 Ti Aorus

Subject: Graphics Cards | December 7, 2017 - 01:31 AM |
Tagged: pascal, GTX 1070Ti, GP104, gigabyte, aorus

Gigabyte is jumping into the custom GTX 1070 Ti fray with the Aorus branded GeForce GTX 1070 Ti Aorus. The new custom graphics card measures 280 x 111 x 38mm and features a WindForce 3 cooler with backplate and a custom 6+2 power phase.

GIGABYTE-AORUS-GTX1070Ti-2.jpg

Backlit by RGB Fusion LEDs, the Aorus logo sits on the side of the card and can be configured to the color of your choice. The shroud is black with orange accents and has sharp stealth angles that is minimal by comparison with other cards. Gigabyte is using a fairly beefy heatsink with this card. Specifically, three 80mm fans push air over a beefy heatsink that consists of three fin stacks connected by four composite heatpipes. Further, the cooler uses direct contact for the heatpipes above the GPU and a metal plate with thermal pads to cover the GDDR5 memory chips. The rightmost fin stack cools the MOSFETs. Additionally, the full cover backplate adds rigidity to the card and has a copper plate to draw excess heat from the underside of the GPU.

The Aorus graphics card is powered by a single 8-pin PCI-E power connector that feeds a 6+2 power phase. External video outputs include one DVI, one HDMI 2.0b, and three DisplayPort 1.4 ports.

The Pascal-based GP104-300 GPU (2432 CUDA cores, 152 TMUs, and 64 ROPs) is clocked at 1607 MHz base and 1683 MHz boost which is the maximum vendors can clock the cards out of the box. Gigabyte does offer 1-click overclocking using its Aorus Graphics Engine software and guarantees at least 88 MHz overclocks to 1683 MHz base and 1771 MHz boost. Users can, of course, use other software like MSI Afterburner or EVGA Precision X if they wish but will need to use the Gigabyte tool if they want the single click automatic overclock. The 8GB of GDDR5 memory is stock clocked at 8008 MHz and sits on a 256-bit bus.

Looking online, the Aorus GTX 1070 Ti Aorus doesn’t appear to be available for sale quite yet, but it should be coming soon. With the Gigabyte GTX 1070 Ti Gaming card coming in at $469, I’m betting the Aorus card with guaranteed overclock will have a MSRP around $500.

Also read:

Source: Gigabyte

EVGA Launches GTX 1070 Ti FTW Ultra Silent Graphics Card

Subject: Graphics Cards | December 6, 2017 - 06:50 PM |
Tagged: evga, ftw, gtx 1070 ti, pascal, overclocking

EVGA is launching a new Pascal-based graphics card with a thicker 2.5 slot cooler in the form of the GeForce GTX 1070 Ti FTW Ultra Silent. The new graphics card has a sleek gray and black shroud with two large black fans in standard ACX 3.0 cooler styling, but with a much thicker cooler that EVGA claims enables more overclocking headroom or a nearly silent fan profile on stock settings.

08G-P4-6678-KR_1.jpg

The GTX 1070 Ti FTW Ultra Silent is powered by two 8-pin PCI-E power connectors that feed a 10+2 power phase and enables the cards 235W TDP (reference TDP is 180 watts). The 2432 Pascal GPU cores are clocked at 1607 MHz base and 1683 MHz boost which aligns with NVIDIA's reference specifications. While there are no guaranteed factory overclock here, EVGA is bundling the dual BIOS card with its EVGA Precision XOC and Precision OC Scanner X software for one-click overclocking that dynamically pushes the clocks up to find the optimal overclock for that specific card. The 8GB of GDDR5 memory is also stock clocked at 8008 MHz. Other features include a backplate, white LEDs, and 2-way SLI support.

Display outputs include one HDMI 2.0b, three DisplayPort 1.4, and one DVI port.

08G-P4-6678-KR_6.jpg

The new FTW series graphics card is available now from the EVGA website for $499.99 and comes with a three year warranty.

The graphics card appears to be rather tall, and I am curious how well the beefier heatsink performs and just how "ultra silent" those fans are! Hopefully we can get one in for testing! The $499.99 MSRP is interesting though because it lines up with the MSRP of the GTX 1080, but with the state of the GPU market as it is the price is not bad and actually comes in about in the middle of where other GTX 1070 Ti cards are at. My guess is they will be snatched up pretty quickly so it's hard to say if it will stay at that price especially on third party sites.

Source: EVGA

More GTX 1070 Ti overclocking

Subject: Graphics Cards | November 7, 2017 - 03:21 PM |
Tagged: pascal, nvidia, gtx 1070 ti, geforce, msi

NVIDIA chose to limit the release of their GTX 1070 Ti to reference cards, all sporting the same clocks regardless of the model.  That does not mean that the manufacturers skimped on the features which help you overclock successfully.  As a perfect example, the MSI GTX 1070 Ti GAMING TITANIUM was built with Hi-C CAPs, Super Ferrite Chokes, and Japanese Solid Caps and 10-phase PWM.  This resulted in an impressive overclock of 2050MHz on the GPU and a memory frequency of 9GHz once [H]ard|OCP boosted the power delivered to the card.  That boost is enough to meet or even exceed the performance of a stock GTX 1080 or Vega 64 in most of the games they tested.

1509608832rtdxf9e1ls_1_16_l.jpg

"NVIDIA is launching the GeForce GTX 1070 Ti today, and we’ve got a custom retail MSI GeForce GTX 1070 Ti GAMING TITANIUM video card to test and overclock, yes overclock, to the max. We’ll make comparisons against GTX 1080/1070, AMD Radeon RX Vega 64 and 56 for a complete review."

Here are some more Graphics Card articles from around the web:

Graphics Cards

 

Source: [H]ard|OCP

You were to bring balance to the price, not leave it in darkness!

Subject: Graphics Cards | November 2, 2017 - 03:03 PM |
Tagged: pascal, nvidia, gtx 1070 ti, geforce

It should come as no surprise to anyone how the GTX 1070 Ti performs, better than a GTX 1070 but not quite as fast as a GTX 1080 ... unless you overclock.  With the push of two buttons Ryan was able to hit 1987 MHz which surpasses your average GTX 1080 by a fair margin.  Hardware Canucks saw 2088MHz when they overclocked as well as memory of  8.9Gbps which pushed the performance past the reference GTX 1080 in many games. Their benchmark suite encompasses a few different games so you should check to see if your favourites are there.

The real hope of this launch was that prices would change, not so much the actual prices you pay but the MSRP of cards both AMD and NVIDIA.  For now that has not happened but perhaps soon it will, though Bitcoin hitting $7000 does not help.

GTX1070TI-5.jpg

"NVIDIA’s launch of their new GTX 1070 Ti is both senseless and completely sensible depending on which way you tend to look at things. The emotional among you are going to wonder why NVIDIA is even bothering to introduce a new product into a lineup that’s more than a year old."

Here are some more Graphics Card articles from around the web:

Graphics Cards

 

Author:
Manufacturer: NVIDIA

Here comes a new challenger

The release of the GeForce GTX 1070 Ti has been an odd adventure. Launched into a narrow window of a product stack between the GTX 1070 and the GTX 1080, the GTX 1070 Ti is a result of the competition from the AMD RX Vega product line. Sure, NVIDIA might have speced out and prepared an in-between product for some time, but it was the release of competitive high-end graphics cards from AMD (for the first time in forever it seems) that pushed NVIDIA to launch what you see before us today.

With MSRPs of $399 and $499 for the GTX 1070 and GTX 1080 respectively, a new product that fits between them performance wise has very little room to stretch its legs. Because of that, there are some interesting peculiarities involved with the release cycle surrounding overclocks, partner cards, and more.

IMG_4944.JPG

But before we get into that concoction, let’s first look at the specifications of this new GPU option from NVIDIA as well as the reference Founders Edition and EVGA SC Black Edition cards that made it to our offices!

GeForce GTX 1070 Ti Specifications

We start with our classic table of details.

  RX Vega 64 Liquid RX Vega 64 Air RX Vega 56 Vega Frontier Edition GTX 1080 Ti GTX 1080 GTX 1070 Ti GTX 1070
GPU Cores 4096 4096 3584 4096 3584 2560 2432 1920
Base Clock 1406 MHz 1247 MHz 1156 MHz 1382 MHz 1480 MHz 1607 MHz 1607 MHz 1506 MHz
Boost Clock 1677 MHz 1546 MHz 1471 MHz 1600 MHz 1582 MHz 1733 MHz 1683 MHz 1683 MHz
Texture Units 256 256 256 256 224 160 152 120
ROP Units 64 64 64 64 88 64 64 64
Memory 8GB 8GB 8GB 16GB 11GB 8GB 8GB 8GB
Memory Clock 1890 MHz 1890 MHz 1600 MHz 1890 MHz 11000 MHz 10000 MHz 8000 MHz 8000 MHz
Memory Interface 2048-bit HBM2 2048-bit HBM2 2048-bit HBM2 2048-bit HBM2 352-bit G5X 256-bit G5X 256-bit 256-bit
Memory Bandwidth 484 GB/s 484 GB/s 410 GB/s 484 GB/s 484 GB/s 320 GB/s 256 GB/s 256 GB/s
TDP 345 watts 295 watts 210 watts 300 watts 250 watts 180 watts 180 watts 150 watts
Peak Compute 13.7 TFLOPS 12.6 TFLOPS 10.5 TFLOPS 13.1 TFLOPS 11.3 TFLOPS 8.2 TFLOPS 7.8 TFLOPS 5.7 TFLOPS
MSRP (current) $699 $499 $399 $999 $699 $499 $449 $399

If you have followed the leaks and stories over the last month or so, the information here isn’t going to be a surprise. The CUDA core count of the GTX 1070 Ti is 2432, only one SM unit less than the GTX 1080. Base and boost clock speeds are the same as the GTX 1080. The memory system includes 8GB of GDDR5 running at 8 GHz, matching the performance of the GTX 1070 in this case. The TDP gets a bump up to 180 watts, in line with the GTX 1080 and slightly higher than the GTX 1070.

Continue reading our review of the GeForce GTX 1070 Ti!

NVIDIA Partners with AWS for Volta V100 in the Cloud

Subject: Graphics Cards | October 31, 2017 - 09:58 PM |
Tagged: nvidia, amazon, google, pascal, Volta, gv100, tesla v100

Remember last month? Remember when I said that Google’s introduction of Tesla P100s would be good leverage over Amazon, as the latter is still back in the Kepler days (because Maxwell was 32-bit focused)?

Amazon has leapfrogged them by introducing Volta-based V100 GPUs.

nvidia-2017-voltatensor.jpg

To compare the two parts, the Tesla P100 has 3584 CUDA cores, yielding just under 10 TFLOPs of single-precision performance. The Tesla V100, with its ridiculous die size, pushes that up over 14 TFLOPs. Same as Pascal, they also support full 1:2:4 FP64:FP32:FP16 performance scaling. It also has access to NVIDIA’s tensor cores, which are specialized for 16-bit, 4x4 multiply-add matrix operations that are apparently common in neural networks, both training and inferencing.

Amazon allows up to eight of them at once (with their P3.16xlarge instances).

So that’s cool. While Google has again been quickly leapfrogged by Amazon, it’s good to see NVIDIA getting wins in multiple cloud providers. This keeps money rolling in that will fund new chip designs for all the other segments.

Source: Amazon