Nvidia Thread

That line of thinking never works on retail world. Either you bring your A game or disappear to irrelevance. If people choose Nvidia over AMD even though they offer competitive product, then they need to do much better job marketing. Don't expect people to buy the product for the sake of.

 

TITAN RTX | I9 9900K vs R7 2700X vs I7 9700K vs I7 8700K | Tested 13 Games |
For Gamers
Published on Dec 23, 2018


NVIDIA Titan RTX Review: Overclocking, Gaming, Power, & Thermals
Gamers Nexus
Published on Dec 22, 2018
For this Titan RTX review, we're looking at thermal performance, noise levels, power consumption, gaming benchmarks, and overclocking.
Article: https://www.gamersnexus.net/hwreviews...
The NVIDIA Titan RTX is a $2500 video card that's outfitted with 24GB of GDDR6, making it most suitable for VRAM-limited tasks on something with RTX 2080 Ti performance. That's really what we're working with, here: If you've wanted a 2080 Ti for some sort of production workload, like CUDA-accelerated rendering, but would be bound by 11GB GDDR6, the Titan RTX would be the alternative. People are still going to buy it for gaming and enthusiast use, though, and so we focused our initial testing on overclocking, gaming, power consumption, and thermals. The Titan RTX will not be made by board partners and is an NVIDIA-only part but, as we saw in our tear-down, it's just a 2080 Ti PCB. This means you could somewhat easily install a 2080 Ti cooler onto the Titan RTX, the "somewhat" mostly being that you'd need to find a 2080 Ti cooler. Waterblocks would work.

Spoiler

 

...That channel is fake.

Look at their fps differences between the 8086K and 8700K which is total non-sense. 10fps difference lol, ya right. This channel has long been known to be fake. I can make a video with made up graphs too. It's sad some people actually subscribe or watch these videos.

 

Alright guys, my video is finally done. Man that was a lot of work. I was so upset about the poor video quality I almost didn't publish it. But, I actually like the 2080 Ti better than I imagined that I would. Notwithstanding the fact that it is stupid expensive, this sucker is awesome. I can hardly wait to see how it runs without a cancer vBIOS (when one is found) and a hard mod to unlock the power and voltage limits.

 

First thing first, I thought you were Hulk Hogan with that new look!
It was really helpful for providing Side by side comparison because midway I forgot all the scores between 1080Ti,2080Ti(stock) and 2080 Ti(chilled water). Truth be told, in most benchmarks your ideal 1080 Ti really shines like a champ. I was shocked to see that 2080Ti chilled water actually pushed the graphics score of Timespy Extreme and FS extreme better than 1080Ti which improved the overall score.
I'm patiently waiting for your benching results on custom vBIOS for EVGA 2080Ti.

 

Thank you. I am flattered. Who doesn't love the Hulkster?

The side-by-side part of the video with graphs was important for anyone wanting to make sense of it. It took me an extra couple of hours to produce that part of the video, but I think doing so was important.

In most things there is not a gigantic difference between my modded 1080 Ti and 2080 Ti running them at stock clocks, and it would be stupid to buy the RTX card just for playing games unless I were building a new system rather than upgrading. But, there was a respectable gap between them in FPS on a couple of the titles I tested.

Yeah, my 1080 Ti is wicked, but it's not a "normal" one by any stretch of the imagination. The gap would have been larger with an unmodded 1080 Ti running ordinary vBIOS. I am sure that if I can figure out how to mod the 2080 Ti the same way it will mop the floor with the 1080 Ti. But, I have already moved up in ranking a TON with just a few benchmarks with the 2080 Ti. It shines much brighter when overclocked that it does running stock against my modded 1080 Ti.

 

It's all Champagne, but bloody hell I wanted to see you open the god damn box! Ha, joking! That EVGA 2080ti is one chunky card, it's like a brick - I like it! I noticed that you gain about a 9% performance uplift from the chilled water on the 2080ti, just at stock (because you haven't compared max overclocks on both), so that's a good boost, I imagine that could widen even more if both were overclocked to the max with custom vBIOS, etc. I was referring to the Superposition benchmark for the 9% difference, and that's one of the most relevant because it's a really long benchmark (heat build up/gaming related) & is for sure GPU bound, and I think it's also the benchmark that shows the cooling in the best light, but again I find it to be the most relevant. Thanks for that hard work in the vid, I can see that was a lot of testing you did, I don't think it matters about the video picture quality. While I was watching your video I was wondering if I'd ever do something like that, in terms of releasing a video review of whatever, and how I'd approach it, kudos for going on camera, I'm sticking with the written word for now, I don't have to obsess about presentation (well, just the words!)!

 

Thanks for the compliments, bro. I appreciate the words of encouragement.

People also have to remember that my 1080 Ti was hard-modded. I was not going to de-solder parts for the video, but the gap would have been wider if I were testing a 1080 Ti that did not have the power limit issue eliminated. That shows up even in gaming and running stock clocks.

Edit: @Robbo99999 I think I mentioned that my initial impression was that the harder the GPU is pushed, the more distance there is between 1080 Ti and 2080 Ti and it doesn't seems like much difference in lighter workloads. That may be more the case than I had first thought. Besides the Superposition benchmark, which is a good indicator, also look at the Deus Ex benchmark with MSAA off versus MSAA 2X. The 1080 Ti cannot hold 60 FPS with MSAA 2X at 1440p.

 

NVIDIA GeForce RTX 2070 Max-Q Performance Leaks Out – Faster Than A Desktop-Grade Radeon RX Vega 64 Graphics Card
https://wccftech.com/nvidia-geforce-rtx-2070-max-q-gpu-performance-benchmark-leak/

 

So how did you make a vbios mod for 1080ti? Have you tried 1080ti and 2080ti in SLI?

Depends on what marketing team took as a reference. If they took Vega 56 on a ryzen laptop then its good otherwise pure desktop AMD cards will demolish Max Q RTX.

 

(That's cool, but in my post I was talking about the comparison between air cooled 2080ti and chilled water 2080ti, when I re-read my post I didn't make it 100% clear.)

 

It's cool. It applies to both scenarios, so it's accurate either way.

 

And Mobile 2080Ti is still not refuted + AMD's best desktop graphics card killed by Nvidia's Mobile Scam.

 

That's an odd comparison in the title... far more people are familiar with the GTX 1080 performance than the Vega 64.

So the 2070 Max-Q is how much slower than a desktop 2070? No comparison is shown.

The 2070 Max-Q is 6.39% faster than the Vega 64 air-cooled. That would make the 2070 Max-Q slower than the Vega 64 LC.

The 2070 Max-Q is 6.50% faster than the desktop 1080, in case you were wondering.

So if you can find a good deal on a desktop Vega 64 or Vega 64 LC, it's a good alternative to the desktop 1080 and desktop 2070, which is what I got out of the comparison chart.

As you can see @Dr. AMK already posted that.

Also the Vega 64 is listed in their chart as slightly faster than the desktop 1080, which makes the Vega 64 LC faster than the 2070 Max-Q in that chart.

So far Mobile RTX GPU's aren't significantly faster than 10 Series mobile / desktop GPU's.

So that add's one more total failure for the Nvidia RTX generation GPU's, yet another Nvidia RTX flop.

 

And normal Mobile is what? 15-17% faster than the Max-Q = +20% faster than desktop 1080

Sorry. Edited with the real deal... The honest text

 

There are no RTX mobile non-Max-Q 2070 / 2080 numbers for comparison. RTX is a new generation so no assumptions can be made as to percentage differences until they are published.

There was a wide range of performance between Series 10 Max-Q GPU's, with some requiring only a small OC to reach stock non-Max-Q numbers, the power, cooling and case fitment made a big difference. Even more so for the 20 series I expect.

What would have helped that article would have been 2070 desktop or 2070 non-Max-Q mobile results in that chart, and without that we can't tell the lameness of the 2070 Max-Q vs 2070 non-Max-Q.

 

I can't go good for this. But this is what Notebookcheck.net show for now...

NVIDIA GeForce RTX 2080 (Laptop)
NVIDIA GeForce RTX 2080 Max-Q)
NVIDIA GeForce RTX 2070 (Laptop)
NVIDIA GeForce RTX 2070 Max-Q)

NVIDIA GeForce GTX 1080 (Laptop)

The Ultimate Scam... The drivers for the Max-Q version were optimized for efficiency (and not performance; only for Max-Q models), there are optimized voltage converters for 1V operation, high-end cooling methods, and a 40 dB limit for the fan noise (with clock adjustments to ensure this at all times). Pay the same or more. Get less!

 

These are the automotive equivalent of RTX 20 series. And, they are "benching" the little Focus (think laptop 2080) in @Papusan weather conditions. ♫ Let it snow, let it snow, let it snow. ♪



This is the automotive equivalent of Max Q...

 

For those that don't need or want the top dog GPU, there is the RTX 2070 from EVGA_Official eBay store: https://www.ebay.com/sch/evga_official/m.html?_nkw=&_armrs=1&_ipg=&_from=



For those that do not want to settle for less... there is one left.

 

Hope that I can build a portable MINI-ITX case with this bad boys, I need some insight to do it with the right components.

 

What CPU? Go EVGA with everything. Excellent global warranty and best service of any other company.

More 3DMark Hall of Fame runs with the 2080 Ti... this is totally destroying my overclocked 1080 Ti benchmarks... by A LOT. When I can get my hands on a good vBIOS that allows overvolting and do the power/shunt mod it is going to be an absolute blood bath, LOL.


@Talon

 

i7 9900K is more than enough for me along with 2080Ti.

 

Yeah, good choice. And, I think it definitely offers the best value. The only time you really gain any leverage from HEDT CPUs is for strictly extreme CPU-heavy tasks and benchmarks that benefit from the higher core count and quad channel memory. For gaming, the 8700K and 9900K quite often beat much stronger Intel HEDT and AMD TR CPUs with more than 8C/16T.

Kind of like a much smaller boxer that is light on his feet can tire out and ultimately beat a much stronger heavy weight as long as he steers clear of the right hook death blow. If the big guy can land a punch then it's lights out with one smack to the skull for the little guy, LOL.

 

For the Mini-ITX, I need to choose the motherboard, cpu cooler, case model, GPU size and the power supply very carefully, the small form factor will not accept any mistakes.

 

I have no experience there. I think brother @TBoneSan might have been researching that. For the PSU, getting more watts than necessary is always better than barely enough or not enough. I am not sure if EVGA even has a mini-ITX mobo available right now.

 

NVLink Titan RTX Benchmarks: Gaming & Power Consumption
Gamers Nexus
Published on Dec 23, 2018
NVLink really just means "SLI," at least, when speaking about these cconfigurations. We benchmarked two Titan RTX cards in SLI vs. 2-way SLI 2080 Tis and 1080 Tis.
Article: pending publication

Spoiler: Details...

This review benchmarks the NVIDIA Titan RTX cards in SLI, or NVLink, in a two-way setup for gaming applications. An additional power test is present for comparison versus a single card. Testing primarily highlights SLI 2080 Ti cards, SLI 1080 Tis, and a single Titan RTX, although plenty of other GPUs are present in the benchmarks. If you're looking for the technical "best" GPU configuration for gaming, this is sort of it -- but that doesn't mean it should be purchased, of course, and that's something we talk about in this video.

 

@19:04 Where does Navi need to perform? RTX 2080 Ti is "70 percent" ahead of Vega 64. AMD so far behind. When they release a 7nm product, Nvidia can respond with a simple process change to a 7nm GPU, and without any architecture change bury them again.

 

It's not a valid question as AMD have already said they aren't looking to compete in the high end GPU area.

It's waste of money for AMD (and Nvidia) to do so for too few people willing to throw away 2080ti / Titan RTX kind of money needlessly. Nvidia is having trouble selling their RTX GPU's, why would AMD foolishly put themselves in the same market losing position?

The AMD Navi GPU rumor's are for 15% faster than Vega 64 / GTX 1080 level GPU @ $250 to give performance to the " masses" as the RX 3080. And, that's a rumor, not reality until it ships.

Seriously, it is silly to me wasting the kind of money it takes to feed the 2080ti / Titan RTX fetish, it's not something I can get behind and recommend to anyone, as it's completely unnecessary.

It's sad to me seeing that such competition encourages similar behavior in others that shouldn't be wasting that kind of money and should be spending it more wisely for themselves and their family.

Nvidia doesn't deserve our money, and Nvidia shouldn't be encouraged to continue to raise prices and deliver RTX-like crap in return for our hard earned money.

Nvidia's business sense is as flawed as their RTX product marketing, and that hasn't been good for their investors either.

And, Nvidia's stock continues to fall:
129.57 USD -5.53 (-4.09%) Official Close 12/21/2018 NAS

 

I won't disagree that RTX 2080 Ti prices are insane, but it's the new reality for now if you want top end GPU performance. Not everyone is going to buy this card, and Nvidia doesn't expect them to either.

That said they're selling as fast as Nvidia can make them. I watch the EVGA website regularly and their 2080 Tis go out of stock nearly as fast as they come into stock. Occasionally a GPU will remain in stock a day or so, but they're gone quickly. Right now not a single EVGA 2080 Ti is available on their website as they sold out again. I won't say the 2080 or 2070 is struggling as I don't care about those, but the 2080 Ti is selling pretty much as fast as Nvidia can turn them out. IMO the 2080 is ehh unless purchased at retail. The 2070 at $500~ is IMO a decent value card and would purchased all day long over anything AMD has. I don't hate AMD GPUs, or AMD. In fact I had my brother purchase an AMD R9 390 a few years back when AMD was actually churing out quality product. AMD GPUs today are a sad joke and it doesn't appear they can right the ship anytime soon. Their Navi GPUs will rely on 7nm it seems, and Nvidia will equally be able to use 7nm to once again bury AMD. AMD has already admitted defeat by saying they're not targeting the high end. That's another way of saying we suck and can't compete with Nvidia. If their 7nm product can't compete with Nvidia today, what will their product look like when Nvidia utilizes that huge process shrink?

 

Back to notebooks. AMD won’t do much for laptops (no MXM and less performance). What’s your advice outside AMD advices? Coming broken BGA laptop with 2050/2060? Aka cheapest possible?

I will go LGA+MXM as usual. My Clevo is soon 4 years and coming Mobile RTX2060 won’t be an good upgrade. A side grade.

 

I think it's pretty clear right now that it's not clear how the RTX mobile GPU's are going to excel over the outgoing 10 series.

I don't see how anything but a couple of the dual GPU capable laptops - costing over $4000-$5000 - are going to deliver the RTX 2080 level mobile GPU.

Existing GTX 1080 laptop's are going to continue to be the top of the available / affordable range of gaming laptops, and if you can get a good price on one of those that's what I would recommend right now.

After the new year and RTX mobile laptops are announced with pricing we'll know better, but even so until they get into owners hands we won't know how they perform - and cool - and if they have substantially higher than GTX 1080 performance.

If the 1080ti couldn't fit in a laptop, then a full performance RTX 2080 with GTX 1080ti performance isn't going to fair better.

Look at the multi-slot coolers on those desktop RTX 2080's, they are even larger than the GTX 1080ti's.

 

What is the TDP for 2080? 215 or near 225W? 200w models shouldn’t be much a problem. Bin better chips for Normal Mobile MXM and we get 1080Ti performance. Should be around there. For me with desktop 980 (skipped 1080) this will be a big upgrade. And it won’t cost 5K with single graphics. I don’t need full package with expensive drives etc.

 

The MXM question is unknown too, whether RTX MXM's if they exist will shoehorn into existing laptops.

The $3000 full 1080 laptops, like the MSI / Alienware single GPU laptops are likely going to have at maximum a full 2070, IDK if they can be grown to cool a full performance 2080.

It's going to be an interesting year in 2019 for desktops, maybe not so with laptops with not much of a GPU performance bump over the 10 series mobile GPU's.

Here are some GPU only power consumption numbers for various desktop GPU's:

https://www.kitguru.net/components/...-rtx-2080-ti-founders-edition-11gb-review/16/

 

I'm kind of liking this. Spanking my modded 1080 Ti pretty bad already. Beating its best scores by a substantial margin in most cases.

http://hwbot.org/submission/4019146_ | https://www.3dmark.com/3dmv/5732953

http://hwbot.org/submission/4019141_ | https://www.3dmark.com/3dmv/5732950

http://hwbot.org/submission/4019083_ | https://www.3dmark.com/3dm11/13075920

http://hwbot.org/submission/4019082_ | https://www.3dmark.com/3dm11/13075914

http://hwbot.org/submission/4019086_ | https://www.3dmark.com/sd/5372973

 

Nvidia: Headed For $100
Dec. 23, 2018 1:40 PM ET
About: NVIDIA Corporation (NVDA), Includes: AMD, INTC
https://seekingalpha.com/article/4229855-nvidia-headed-100

" Summary
Nvidia shares have plunged recently, off the back of an utterly disastrous fiscal Q3.

The weakening cryptocurrency market, previously downplayed by management, had a very real effect on revenue, and the inventory issues created by mismanagement will have a material effect over the coming quarters.

Coupled with slowing datacenter growth, a downturn in the overall chip space, and high powered competition (AMD, Intel, in-house), the Nvidia bull thesis is flawed.

The stock may not have significant upside catalyst until 2H of calendar 2019, when Nvidia is expected to fully work through its inventory reset. By this time, the stock markets and global economy may be increasingly volatile.

Without fundamental upside catalysts, slowing growth, and increased competition, the stock likely has more room to fall. Nvidia's stock remains a no-fly zone. Slashing PT from $232 to $100.

Spoiler: Details...

Crypto Bubble Hit Nvidia Harder Than Expected
After the fiscal Q3'2019 report, the majority of the sell-side community was utterly shocked by the size of Nvidia's (NASDAQ: NVDA) miss. Nvidia has had a phenomenal track record of delivering beat-and-raise quarters.

• Revenue of $3.18 billion vs. consensus estimate of $3.24 billion
• EPS of $1.84 vs. consensus estimate of $1.71
• Revenue guide of $2.7 billion (+/- 2%) vs. consensus of $3.4 billion
• Gaming revenue of $1.76 billion vs. consensus of $1.89 billion
• Datacenter revenue of $792 million vs. consensus of $821 million

The real miss was on the gaming side, coming in nearly 7% lower than expectations. The cryptocurrency bubble continuously downplayed by management actually hit Nvidia harder than expected, with revenues falling far below guidance.

Unfortunately for the bulls, because management overestimated the demand within the GeForce lineup, it overproduced. This created an inventory glut. This glut, according to management, will take multiple quarters to work through.

While the entire market was focused on the effect a crypto bubble would have on AMD (NASDAQ: AMD), the market underestimated (partly in thanks to management's comments) the effect lowered crypto demand would have on sales.

Now, despite the fact that Nvidia has come to market with its RTX graphics card lineup, it is going to have inventory issues, particularly with the sales of GTX 1060, 1070, and 1080 cards. These were some of the more popular GPU choices from Nvidia in the crypto mining realm.
Bitcoin Close Price data by YCharts
AMD data by YCharts
NVDA PE Ratio (Forward) data by YCharts
NVDA PS Ratio (Forward) data by YCharts

Nvidia's 54 percent plunge this quarter makes it the biggest loser in S&P 500

• Nvidia shares surged from 2016 through much of 2018 on demand for processors that could support cryptocurrencies and artificial intelligence.
• The momentum is gone, and in the fourth quarter Nvidia has been the worst-performing stock in the S&P 500.

Jordan Novet | @jordannovet, Published 4:05 PM ET Fri, 21 Dec 2018
https://www.cnbc.com/2018/12/21/nvidia-is-the-worst-performing-sp-500-stock-this-quarter.html

 

Msi will get temp problems with 8 core (perfom as 8 cores should) and same TDP MXM graphics as they use today. And we all know Alienware’s status. Massive crippled as today (thin models will get the most crippled versions from what I heard from Nvidia). Can’t see these tragedies will be refresh with thicker chassis.

 

It's done! 4K 144hz @ Ultra Settings! Merry Christmas
Submitted 14 hours ago by ring0r
https://www.reddit.com/r/nvidia/comments/a8ygj6/its_done_4k_144hz_ultra_settings_merry_christmas/


Annoying_Arsehole 41 points 13 hours ago
"Did your pet unicorn have diarrhea?"

ring0r[ S] 16 points 14 hours ago
"Specs: I9 9900K 2x 2080ti 2x Samsung 970 Pro 4x 8GB Crosair V Pro 4000Mhz Seasonic 1300w Prime Gigabyte Z390 Master Acer Predator X27 4k 144hz HDR G-Sync"

 

@Dr. AMK

This might interest you. Nothing we've never seen but this link is from my build which is about 1 year old now. My 7700k is pretty weak sauce now so I'm considering going 9900k + z390 soonish. I think I just want to build something new and I kind of enjoy having limitations and navigating the compromises of a SFF to a degree.
With the delidded 7700k in there at 5ghz the CPU never gets over 71 degrees with a extremely lazy and silent fan profile. That's been enough incentive for me to want to cram more in.

Also a 800 SFX exists now which is pretty neat and something you may want as brother Fox mentioned

Also check out the DAN Case A4-SFX v2 which is pretty neat looking too but you'll be limited to a 92mm AIO which may be a bit testing for a 9900k... if I were to guessing man.

 

The card is two slots, so that should be good in most Mini-ITX cases. See if you can find out the length, that should help narrow down your card selection. For PSU, Silverstone makes a 800W SFX, so you can go that route if needed. If you want to go with Intel HEDT in a mini-ITX, then AsRock is the only player in town at the time of this post. Also, it's a little dated, but most of this is in the Might-Mini ITX thread.

 

Thank you my friends for your kind insights, I really appreciate your support.

 

That was a funny comment. I kind of agree. If it were all one color (especially white for me) it would look pretty sharp. The rainbow is a bit much for my taste. I am more into the monochromatic look. If all of the lights are white, I don't even care about RGB except that it does have some advantages for resale because some people are really crazy-go-nuts about RGB.

 

The TRUTH About RTX Graphics Cards - Ray Tracing & DLSS BENCHMARKED.
PC Centric
Published on Dec 24, 2018
RTX ON or OFF? Ray Tracing and DLSS games are finally here, with Battlefield V and RTX gameplay benchmarks on the 2080 Ti, 2080 and 2070! Join PC Centric for the Nvidia RTX Graphics card review...
Links to everything featured below!


CzM _1 4 hours ago
"I'm trying to break out of my console gaming hovel in to the palace of PC gaming but I'm finding it increasingly harder with price tags this high on GPUs nowadays. It's insanely frustrating."

Alessandro Borges 3 hours ago
"DLSS is just a fancy name for upscaling with some anti-aliasing (FXAA ?) [Old] GTX cards can render 4K content using 1440p rendering and then upscaling it, either with GPU or monitor upscaling (check your Nvidia Control Panel). That's is how I play War Thunder in my Samsung 55ku6000 using a GTX 1070!"

One Cosmic 2 hours ago
"Every youtuber should just blame nvidia for milking people and not do any reviews about dlss or dxr..."

 

RTX 2070 vs VEGA 64 Test in 8 Games
Testing Games
Published on Dec 25, 2018
RX VEGA 64 8GB vs RTX 2070 8GB in 1440p (i7 8700k)

Spoiler: Games :

Forza Horizon 4
Battlefield 5 - 04:05
Assassin's Creed Odyssey - 02:07
Shadow of the Tomb Raider - 03:20
Grand Theft Auto V - 04:52
HITMAN 2 - 06:27
Far Cry 5 - 07:58
The Witcher 3 - 09:08

System:
Windows 10 Pro
Intel i7 8700k 3.7Ghz
GeForce RTX 2070 8Gb
Radeon RX VEGA 64 8Gb
16Gb RAM 3200Mhz

 

https://www.newegg.com/Product/Product.aspx?Item=N82E16814126245&Description=vega 64&cm_re=vega_64-_-14-126-245-_-Product

https://www.newegg.com/Product/Product.aspx?Item=N82E16814487412&Description=rtx 2070&cm_re=rtx_2070-_-14-487-412-_-Product -- Yes this is an overclocked "A" chip.

The Vega is topped out on clocks, the 2070 has another 200-300mhz in the tank as well as another +1000mhz on the vRAM which helps significantly on these 20 series cards. Vega is a hot piece of trash that sucks down power.



The benchmark guy himself found RTX 2070 10% faster on average than the Vega64 at 10:00. Cheaper price and better card in performance, far less power consumption and better thermals. It's an easy choice, RTX 2070 is a far better deal. Vegarbage 64 is a bad deal.

 

NVIDIA Rolls Out Holiday Gift for Pascal Cards Owners – OC Scanner Now Available in MSI Afterburner


----------------------------------

NVIDIA GeForce RTX 2060 to Ship in Six Variants Based on Memory Size and Type Techpowerup.com | Today, 05:39

NVIDIA drew consumer ire for differentiating its GeForce GTX 1060 into two variants based on memory, the GTX 1060 3 GB and GTX 1060 6 GB, with the two also featuring different GPU core-configurations. The company plans to double-down - or should we say, triple-down - on its sub-branding shenanigans with the upcoming GeForce RTX 2060. According to VideoCardz, citing a GIGABYTE leak about regulatory filings, NVIDIA could be carving out not two, but six variants of the RTX 2060!

There are at least two parameters that differentiate the six (that we know of anyway): memory size and memory type. There are three memory sizes, 3 GB, 4 GB, and 6 GB. Each of the three memory sizes come in two memory types, the latest GDDR6 and the older GDDR5. Based on the six RTX 2060 variants, GIGABYTE could launch up to thirty nine SKUs. When you add up similar SKU counts from NVIDIA's other AIC partners, there could be upward of 300 RTX 2060 graphics card models to choose from. It won't surprise us if in addition to memory size and type, GPU core-configurations also vary between the six RTX 2060 variants compounding consumer confusion. The 12 nm "TU106" silicon already has "A" and "non-A" ASIC classes, so there could be as many as twelve new device IDs in all! The GeForce RTX 2060 is expected to debut in January 2019.

 

There are new Radeon RX Vega 64's @ ~$400 if you search around a bit, and even more just over $400 to $499. You can find much higher priced units for sale, but the same goes for any GPU with pre-crypto-crash pricing still listed.

There are new and used RX Vega 64's for under $400 up to $500 on ebay:
https://www.ebay.com/sch/i.html?_oaa=1&_dcat=27386&_udlo=0&_sop=15&_udhi=500&LH_ItemCondition=2000|2500|3000|1500|1000&_fsrp=1&_nkw=amd+radeon+vega+64&_sacat=0&_from=R40&rt=nc

The big win with the AMD RX Vega 64, is it's not an Nvidia product, and even better it's not an RTX product.

You're quoting performance from drivers back in October, my post is using the latest Radeon Adrenaline 2019 + driver updates software; that Hardware Unboxed video's performance numbers are out of date.

The performance gap has narrowed considerably, and the sheer joy of running an AMD CPU + an AMD GPU will far outweigh any small FPS differences.

The combo of an AMD 2700x + an AMD Vega 64 will spread the joy of the Holidays year round. Or, even an RX 580 / RX 590 build would be nice.

Note: The single fan AMD RX 64 GPU's will be *loud* at full load, so make sure you have a large case (not a small case) with good air flow, and/or spend more $ for a dual-fan or tri-fan model.

 

And how many of those used RX Vega 64's is used for mining?

 

Probably far fewer AMD Vega 64's than the number of Nvidia 1080ti's / 1080's / 1070's / 1060's used as mining GPU's.

It's not necessarily a bad thing to have been a mining card as most miners know to tune their GPU's for undervolt to reduce electricity costs - it's part of the set up for optimizing profits.

There's actually a lot of new in box GPU's coming from mining sites that never got time to put them into service before mining for crypto became unprofitable.

Like this ebay seller that has sold 57 new in box units, and has 5 left:

Sapphire AMD Radeon RX Vega 64 8GB Vega64 8G OC HBM2 New in The Box - $449 - BuyItNow

There are real bargains to be had from mining firms dumping hardware as fast as they can to pay debts incurred during the boom.

Here's the lowest priced 3 fan model:

PowerColor AMD Radeon RX Vega 64 RED DEVIL 8GB HBM2 GDDR5 PCI Express 3 - $475 - BuyItNow

And, there are others...

And, Vega 56's...

 

http://forum.notebookreview.com/threads/nvidia-thread.806608/page-14#post-10678525

And from a post below in same site... “So again, 1/10 MSRP should be the most we pay for a discarded GPU from a mining operation”.

What is RX64 MSRP?

 

Apologies. Pls delete.