Ask GN 75 is an excellent episode. We had great questions for this one, including discussion on X370 vs. X470 benchmarking for Ryzen 2000 series CPUs (e.g. R7 2700X, R5 2600X), which we’ll get in to more detail with in the near future. As noted in the episode, we’re technically not under embargo for the Ryzen 2 CPUs, but we’re planning to hold our review until embargo lift out of respect for AMD’s decision to stop giving special treatment to some media, for this round. That said, we still talk a bit about X370 vs. X470 benchmarking in the Ask GN episode.
The other excellent topic pertained to receiving review samples and balancing hardware criticism – basically behind-the-scenes politics. Find the episode below:
Intel has migrated its new i9 processor family to the portable market, or semi-portable, anyway. The i9-8950HK is an unlocked, overclockable 6C/12T laptop CPU, capable of turbo boosting to 4.8GHz when power and thermal budget permits.
The new i9-8950HK runs at a base clock of 2.9GHz, with single-core turbo boosting to 4.8GHz. We are unclear on what the all-core turbo boost max frequency is, but it’s clearly lower. TDP is rated at 45W, though note that this is more a measure of the cooling requirements and not the actual power consumption; that said, Intel’s TDP ratings almost always explicitly coincide with power consumption numbers. The 8950HK will move away from the quad-channel support of its desktop brethren, and instead move down to dual-channel memory at a base frequency of 2666MHz. Unlocked motherboards should theoretically permit higher memory speeds, though we are uncertain of the market options at this time.
Revealed to press under embargo at last week’s GTC, the nVidia-hosted GPU Technology Conference, nVidia CEO Jensen Huang showcased the new TITAN W graphics card. The Titan W is nVidia’s first dual-GPU card in many years, and comes after the compute-focused Titan V GPU from 2017.
The nVidia Titan W graphics card hosts two V100 GPUs and 32GB of HBM2 memory, claiming a TDP of 500W and a price of $8,000.
“I’m really just proving to shareholders that I’m healthy,” Huang laughed after his fifth consecutive hour of talking about machine learning. “I could do this all day – and I will,” the CEO said, with a nod to PR, who immediately locked the doors to the room.
NVidia today announced what it calls “the world’s largest GPU,” the gold-painted and reflective GV100, undoubtedly a call to its ray-tracing target market. The Quadro GV100 combines 2x V100 GPUs via NVLink2, running 32GB of HBM2 per GPU and 10,240 CUDA cores. NVidia advertises 236 TFLOPS Tensor Cores in addition to the power afforded by the 10,240 CUDA cores.
Additionally, nVidia has upgraded its Tesla V100 products to 32GB, adding to the HBM2 stacks on the interposer. The V100 is nVidia’s accelerator card, primarily meant for scientific and machine learning workloads, and later gave way to the Titan V(olta). The V100 was the first GPU to use nVidia’s Volta architecture, shipping initially at 16GB – just like the Titan V – but with more targeted use cases. NVidia's first big announcement for GTC was to add 16GB VRAM to the V100, further adding a new “NV Switch” (no, not that one) to increase the coupling capabilities of Tesla V100 accelerators. Now, the V100 can be bridged with a 2-billion transistor switch, offering 18 ports to scale-up the GPU count per system.
A few days ago, we ran our most successful, highest-watched livestream in the history of GN. The stream peaked at >5300 concurrent viewers for around 2.5 hours, during which time we attempted to outmatch the LinusTechTips 3DMark score submitted to the 3DMark Hall of Fame. This was a friendly media battle that we decided to bring to LTT after seeing their submission, which briefly knocked us out of the Top 10 for the Hall of Fame. As noted in this recap video, we're not skilled enough to compete with the likes of K|NGP|N, dancop, der8auer, or similar pro XOCers, but we can certainly compete with other media. We made a spectacle of the event and pushed our i9-7980XE, our RAM, and our GPU (a Titan V) as far as our components would allow under ambient cooling. Ambient, by the way, peaked at ~30C during the stream; after the stream ended and room ambient dropped ~10C to 20C, our scores improved to 8285 in Timespy Extreme. This pushed us into 4th place on the 3DMark official Hall of Fame, and 3rd place in the HW Bot rankings.
The overclocking stream saw guest visits from Buildzoid of Actually Hardcore Overclocking, who assisted in tuning our memory timings for the final couple of points. We think there's more room to push here, but we'd like to keep some in the tank for a retaliation from Linus and team.
If, to you, the word "unpredictable" sounds like a positive attribute for a graphics card, ASRock has something you may want. ASRock used words like “unpredictable” and “mysterious” for its new Phantom Gaming official trailer, two adjectives used to describe an upcoming series of AMD Radeon-equipped graphics cards. This is ASRock’s first time entering the graphics card space, where the company’s PCB designers will be faced with new challenges for AMD RX Vega GPUs (and future architectures).
The branding is for “Phantom” graphics cards, and the first-teased card appears to be using a somewhat standard dual-axial fan design with a traditional aluminum finstack and ~6mm heatpipes. Single 8-pin header is shown in the rendered teaser card, but as a render, we’re not sure what the actual product will look like.
The past week of hardware news has been peculiarly busy for this time of year, with a deluge of news posting toward the latter half of last week. For major stories, [H]ardOCP’s coverage of nVidia’s GPP agreements has undoubtedly garnered among the most attention in the news cycle, with additional stories of interest covering hacks to get Coffee Lake CPUs functional in Z170 and Z270 motherboards.
We’ve got a couple of minor news items – new liquid coolers, a mini-review of a chair – and a couple of game industry items, like Valve’s return to game development.
Find the written and filmed recaps below:
We're back with Ask GN! It's been a long week of testing: Patrick has been working on FFXV, an Elgato 4K60 review, and other pieces; I've been working on managing the upcoming travel schedule, primarily for Computex and other tradeshows, and also have a whole slew of in-depth content coming up. One of our biggest endeavors for the week will be our upcoming livestream, where we intended to battle the LinusTechTips team for a top 10 spot in 3DMark benchmark rankings. It's a bit of a friendly rivalry, and we think you all will enjoy tuning in. We'll talk about that more soon.
We also have a news video going up tomorrow, as usual. The video will include several major news items for the past week, including some discussion of the nVidia GPP story that's been going around. Stay tuned for all of that.
In the meantime, Ask GN is below, and the timestamps are below that. Our Patreon bonus episode is here.
01:37 - David Watson: “Hey Steve , do you think we will end up seeing Nvidia competing with the aftermarket cards more directly by releasing their own non founders edition cards with new custom cooling solutions and heatsinks with double or triple fan designs ? i think its a fascinating prospect and one i feel Nvidia has had a lot of thought bent on and is surely bound to use much sooner rather than later because if they can get more marketshare then they surely will and i honestly feel its coming from Nvidia and could be massive for them , because i know one thing , if they released a badass new aftermarket card design that not only performed really well but looked really cool and actually ran cool then i for one would certainly be very tempted by it as many surely would Steve , do you feel this is on the horizon ? thanks man”
06:47 - Stank Buddha: “Quick question, regarding the 200fps limit in (newer?) games. Is this applied per monitor if you were using multi monitor or is it the actual game that is locked to to pushing 200fps total??? Like if the game is locked itself then a theoretical 3 60hz monitors would be maxing it out(60*3=180). Or can you do a 3 monitor 240hz and max em out each at 200fps. just wondering.”
08:57 - Michael Morgan: “Can you demonstrate the end user benefit of HBM memory over GDDR5 or GDDR5X on GPU's please?”
13:12 - vishal bobde: “#askgn-questions Why do CPU don't have different manufacturers like GPU. If there were more manufacturers we might get more enthusiasts features from factory like LM tim and better IHS.”
17:23 - Satoshi_Nakamoto: “@GN Staff Hey guys could you reach out to Thermaltake and ask them if they have any idea for the arrival of the Level 20 Case?”
18:14 – defenestrationize: “Steve, a massive limit limit for APUs is their need to use system memory. Do you think, APUs will remain on the low end or end up high end (is there not actually a limit to DDR for APUs) , More memory channels on APUs (possibly separate for CPU/GPU so 2 sticks DDR 2 sticks GDDR) or on chip memory (hbm) will appear in the near future? Given how board partners operate and push for chip consolidation , do you think we might see a MB, RAM, GPU, CPU as a single pcb ? Feel free to cut this question as needed to perhaps a simpler version.”
22:36 - Dayne_ofStarfall: “@Steve Burke Hello Steve, I’m a bit confused by case fans lately, specifically RPM and in relation to voltage. If I understand correctly different fans have different MAX and MIN RPM at a given voltage. But what happens when you connect two fans with different MIN/MAX RPM to a single header on the motherboard using a Y-splitter? Do the fans spin at different RPMs? And how does this work when they’re connected to a SATA-powered PWM Hub (like the one that comes with most Phanteks cases)? Also what is the amount of fans that can be safely connected to one header? I’ve read on forums that the cable or port can catch fire if they draw too much power, is this true? Thank you.”
24:25 - Ash_Borer: “#askgn-questions how do delidded (with LM) temperatures compare to soldered CPU temperatures? Do you ever plan to delid a ryzen and test the results? Im under the assumption that delidding provides better temps, so i dont mind that intel doesnt solder anymore - as an enthusiast i want to delid anyway and if its not soldered it is easier to delid.”
25:29 - Armand B.: “Modmat out of stock ? DAMN MINERS !”
25:53 - Nory The Explorer.exe: “@GN Staff What is an important fact, viewers should know about GamersNexus?(edited) And the opposite, what is a big misconception viewers have expressed about GamersNexus?”
Host: Steve Burke
Video: Andrew Coleman
Khronos Group today released the Vulkan 1.1 and SPIR-V 1.3 updates. Adoption of both Vulkan and DX12 has been limited, so the overall purpose of this update is described as “Building Vulkan’s Future.”
“Intel & NVIDIA working together” will surely raise eyebrows, but this isn’t similar to how AMD and Intel recently worked to make Hades Canyon. Rather than work together on a product, the two companies sent high-ranking researchers and engineers to meet with the US Government, alongside numerous other AI and machine learning organizations. Intel and nVidia are the most relevant to our line of work, and the representatives at each organization worked to educate government officials on the needs of AI and machine learning development.
In addition to this news, Intel also is working on a $5B expansion for 10nm production, covered further down, and the company’s new Coffee Lake Pentium CPUs have been rumored and assigned specs.
Show notes below the video, if you prefer to read.
We moderate comments on a ~24~48 hour cycle. There will be some delay after submitting a comment.