• Welcome to TechPowerUp Forums, Guest! Please check out our forum guidelines for info related to our community.

Water Cooled Vega 64 Overclocked And Tested By TomsHardware

Joined
Sep 17, 2014
Messages
5,042 (3.67/day)
Likes
4,041
Location
Duiven, Netherlands
Processor i7 8700k 4.8Ghz @ 1.31v
Motherboard AsRock Fatal1ty K6 Z370
Cooling beQuiet! Dark Rock Pro 3
Memory 16GB Corsair Vengeance LPX 3200/C16
Video Card(s) MSI GTX 1080 Gaming X @ 2100/5500
Storage Samsung 850 EVO 1TB + Samsung 830 256GB + Crucial BX100 250GB + Toshiba 1TB HDD
Display(s) Eizo Foris FG2421
Case Fractal Design Define C TG
Power Supply EVGA G2 750w
Mouse Logitech G502 Protheus Spectrum
Keyboard Sharkoon MK80 (Brown)
Software W10 x64
#76
No, it's you people who babble how GCN is all the same from HD7000 up to Vega that piss me off. And now all of a sudden my absence of owning an AMD is now a "problem" because I'm supposedly not knowledgeable enough about GCN even though I freaking owned the very first generation of it on excellent HD7000 series. Weren't you among the idiots who were yelling at me for being an AMD fanboy coz I had a fucking Radeon 2 years ago before GeForce, but now, all of a sudden, it's an issue coz I supposedly don't know anything about it coz I don't have it now. Consistency, you may want to look that up in dictionary...

Also, that 4 wheels and steering wheel car analogy. You may want to study that up as well...
Come on man, I've also been putting that argument in and Vega's current state of affairs shows exactly what this radically changed (sarcasm, I hope you can see) GCN does for its efficiency (zero) its clockability (no real change from Polaris) and its real life performance (all over the place from below GTX 1070 to 10-15% over 1080). Essentially, the practical results of this new iteration of GCN are exactly as they were before and even WITH the die shrink, it has lost efficiency compared to Fury X because AMD had to push heavy on the power budget to achieve reasonable clocks.

If you take a few steps back you can see comparable things happening with Intel's Core arch and AMD's GCN - they're both architectures that are closing in on their EOL date, and are extended by adding more juice (or cores!) to achieve higher absolute performance. Nvidia does not follow that line - every iteration of CUDA has tangible improvements, and every consumer GPU from Kepler onwards pushed an efficiency jump by tearing down every barrier to enable ever higher clocks at the expense of versatility. From a 100-150mhz gap at the Kepler - Kepler Refresh (already!) this has grown to 200-450 (!) mhz gaps in Pascal compared to GCN. That's a LOT of time for a competitor to follow suit and make wise decisions - and Vega is the result of not doing just that. Instead, AMD markets a 'prosumer' Vega as a jack-of-all-trades, it couldn't have been more ironic.

I remember you harping on about tiled rasterization and all this other new stuff, but it simply does not translate to performance or efficiency at all. That goes a long way to prove RTG's skill at redesigning something aren't really very good, and it supports what most have been saying that the RTG press slides were a bit too optimistic, as always. Its the good old pitfall of over theorizing something that is about to happen but never really does happen; in the same vein, look at where that awesome DX12 GCN performance is going compared to Pascal - again we see higher CPU overhead in the driver... Again it underlines that GCN is still a lumbering giant and not a lean machine - and that's 2012 knowledge; five full years of a practical standstill.
 
Last edited:
Joined
Jan 6, 2014
Messages
73 (0.04/day)
Likes
62
Location
Germany
System Name Main Machine
Processor Intel i7-7700K
Motherboard ASUS ROG MAXIMUS IX CODE
Cooling Water cooling, EK-DDC 3.2 PWM, 2x240mm+1x120mm Rads, CPU XSPC RayStorm
Memory 16GB G.Skill RipJaws DDR4-3200
Video Card(s) Sapphire R9 Radeon Vega 64
Storage Samsung SSD 960 EVO NVMe 250GB, Samsung SSD 840 EVO 1TB, Samsung SSD 840 Pro 256GB,4x3TB Seagate
Display(s) BenQ ZOWIE XL2730 144Hz 27 inch Freesync
Case Cooler Master Storm Trooper with side window
Audio Device(s) Turtle Beach Elite Pro Tournament + Elite Pro TAC
Power Supply be quiet! Dark Power Pro 10 1200W
Mouse Logitech G900 Chaos Spectrum
Keyboard ROCCAT Ryos MK Pro
Software Win 10 Pro
#77
Yesterday night I finished building the custom water loop on my Vega 64.
I started building stuff late in the evening and it got quite late until 2:00 o'clock.

I did only one test so far. I played Witcher Wild Hunt for around 40 minutes leaving the Wattman settings at default Balanced mode.
I was able to achieve average clock of around 1528MHz with the GPU temperature staying on average at 39°C at room temperature of 24°C. These temperatures are great. :peace:
The maximum temperature of the GPU was 42°C. The maximum GPU frequency was 1630MHz.
The frequency measurements were done with Wattman as well as GPU-Z.
I will repeat the test tonight and make some screen shots as I did not make any yesterday.

This result is very close to what Tom's got in the Witcher with Turbo mode.
In my case the Wattman is on balanced mode and I achieve average of 1528MHz in Witcher.
I will do further testing tonight with Turbo mode turned on to see how high the average clock can go on my Vega 64 sample.

I bought a Saphire aircooled Vega 64. It is a variant where the GPU chip and HBM2 have same height and there is Epoxy filling between GPU and HBM2.
I used IC Diamond 7 Carat as thermal compound on GPU and HBM.

The undervolting test will come in the next days.

With original cooler it was hitting 85°C all the time with default fan profile.
To achieve lower temperatures in range of 65°C on the blower cooler, the fan speed needed to be increase above 3300RPM.
It was like sitting next to a Vacuam cleaner!!
The average clock there with high fan RPM was always below 1430MHz.
 
Last edited:
Joined
Mar 10, 2010
Messages
5,371 (1.78/day)
Likes
1,848
Location
Manchester uk
System Name Quad GT evo V
Processor FX8350 @ 4.6ghz1.435c NB2.44ghz Ht2.64ghz
Motherboard Gigabyte 990X Gaming
Cooling 360EK extreme 360Tt rad all push/pull, cpu,NB/Vrm blocks all EK
Memory Corsair vengeance 16Gb @1600 cas8
Video Card(s) Rx vega 64 waterblockedEK
Storage samsung 840(250OS), WD 1Tb+2Tb +3Tbgrn, 1tb seagate sshd hybrid(games)
Display(s) Samsung uea28"850R 4k freesync, LG 49" 4K 60hz ,Oculus
Case Custom(modded) thermaltake Kandalf
Audio Device(s) Xfi creative 7.1 on board ,Yamaha dts av setup
Power Supply corsair 1200Hxi
Mouse CM optane
Keyboard CM optane
Software Win 10 Pro
Benchmark Scores 15.69K best overall sandra so far 6600 3dmark
#78
Yesterday night I finished building the custom water loop on my Vega 64.
I started building stuff late in the evening and it got quite late until 2:00 o'clock.

I did only one test so far. I played Witcher Wild Hunt for around 40 minutes leaving the Wattman settings at default Balanced mode.
I was able to achieve average clock of around 1528MHz with the GPU temperature staying on average at 39°C at room temperature of 24°C. These temperatures are great. :peace:
The maximum temperature of the GPU was 42°C. The maximum GPU frequency was 1630MHz.
The frequency measurements were done with Wattman as well as GPU-Z.
I will repeat the test tonight and make some screen shots as I did not make any yesterday.

This result is very close to what Tom's got in the Witcher with Turbo mode.
In my case the Wattman is on balanced mode and I achieve average of 1528MHz in Witcher.
I will do further testing tonight with Turbo mode turned on to see how high the average clock can go on my Vega 64 sample.

I bought a Saphire aircooled Vega 64. It is a variant where the GPU chip and HBM2 have same height and there is Epoxy filling between GPU and HBM2.
I used IC Diamond 7 Carat as thermal compound on GPU and HBM.

The undervolting test will come in the next days.

With original cooler it was hitting 85°C all the time with default fan profile.
To achieve lower temperatures in range of 65°C on the blower cooler, the fan speed needed to be increase above 3300RPM.
It was like sitting next to a Vacuam cleaner!!
The average clock there with high fan RPM was always below 1430MHz.
Similar ideas to me but im cash strapped so the waterblocks on hold , please consider joining the vega club in the club section , its Getting lonely in there.
 
Joined
Aug 6, 2009
Messages
677 (0.21/day)
Likes
154
Location
Chicago, Illinois
#79
Nic
Yesterday night I finished building the custom water loop on my Vega 64.
I started building stuff late in the evening and it got quite late until 2:00 o'clock.

I did only one test so far. I played Witcher Wild Hunt for around 40 minutes leaving the Wattman settings at default Balanced mode.
I was able to achieve average clock of around 1528MHz with the GPU temperature staying on average at 39°C at room temperature of 24°C. These temperatures are great. :peace:
The maximum temperature of the GPU was 42°C. The maximum GPU frequency was 1630MHz.
The frequency measurements were done with Wattman as well as GPU-Z.
I will repeat the test tonight and make some screen shots as I did not make any yesterday.

This result is very close to what Tom's got in the Witcher with Turbo mode.
In my case the Wattman is on balanced mode and I achieve average of 1528MHz in Witcher.
I will do further testing tonight with Turbo mode turned on to see how high the average clock can go on my Vega 64 sample.

I bought a Saphire aircooled Vega 64. It is a variant where the GPU chip and HBM2 have same height and there is Epoxy filling between GPU and HBM2.
I used IC Diamond 7 Carat as thermal compound on GPU and HBM.

The undervolting test will come in the next days.

With original cooler it was hitting 85°C all the time with default fan profile.
To achieve lower temperatures in range of 65°C on the blower cooler, the fan speed needed to be increase above 3300RPM.
It was like sitting next to a Vacuam cleaner!!
The average clock there with high fan RPM was always below 1430MHz.[/QUOTE
Yesterday night I finished building the custom water loop on my Vega 64.
I started building stuff late in the evening and it got quite late until 2:00 o'clock.

I did only one test so far. I played Witcher Wild Hunt for around 40 minutes leaving the Wattman settings at default Balanced mode.
I was able to achieve average clock of around 1528MHz with the GPU temperature staying on average at 39°C at room temperature of 24°C. These temperatures are great. :peace:
The maximum temperature of the GPU was 42°C. The maximum GPU frequency was 1630MHz.
The frequency measurements were done with Wattman as well as GPU-Z.
I will repeat the test tonight and make some screen shots as I did not make any yesterday.

This result is very close to what Tom's got in the Witcher with Turbo mode.
In my case the Wattman is on balanced mode and I achieve average of 1528MHz in Witcher.
I will do further testing tonight with Turbo mode turned on to see how high the average clock can go on my Vega 64 sample.

I bought a Saphire aircooled Vega 64. It is a variant where the GPU chip and HBM2 have same height and there is Epoxy filling between GPU and HBM2.
I used IC Diamond 7 Carat as thermal compound on GPU and HBM.

The undervolting test will come in the next days.

With original cooler it was hitting 85°C all the time with default fan profile.
To achieve lower temperatures in range of 65°C on the blower cooler, the fan speed needed to be increase above 3300RPM.
It was like sitting next to a Vacuam cleaner!!
The average clock there with high fan RPM was always below 1430MHz.
Great results, thanks for sharing. All settings on high? Resolution?
 
Joined
Jan 6, 2014
Messages
73 (0.04/day)
Likes
62
Location
Germany
System Name Main Machine
Processor Intel i7-7700K
Motherboard ASUS ROG MAXIMUS IX CODE
Cooling Water cooling, EK-DDC 3.2 PWM, 2x240mm+1x120mm Rads, CPU XSPC RayStorm
Memory 16GB G.Skill RipJaws DDR4-3200
Video Card(s) Sapphire R9 Radeon Vega 64
Storage Samsung SSD 960 EVO NVMe 250GB, Samsung SSD 840 EVO 1TB, Samsung SSD 840 Pro 256GB,4x3TB Seagate
Display(s) BenQ ZOWIE XL2730 144Hz 27 inch Freesync
Case Cooler Master Storm Trooper with side window
Audio Device(s) Turtle Beach Elite Pro Tournament + Elite Pro TAC
Power Supply be quiet! Dark Power Pro 10 1200W
Mouse Logitech G900 Chaos Spectrum
Keyboard ROCCAT Ryos MK Pro
Software Win 10 Pro
#80
Nic


Great results, thanks for sharing. All settings on high? Resolution?
Yes all the settings except for the Nvidia Hairwork are at high.
I am using Benq Zowie XL2730 QHD monitor at default 2560 x 1440 resolution.
 
Joined
Jan 6, 2014
Messages
73 (0.04/day)
Likes
62
Location
Germany
System Name Main Machine
Processor Intel i7-7700K
Motherboard ASUS ROG MAXIMUS IX CODE
Cooling Water cooling, EK-DDC 3.2 PWM, 2x240mm+1x120mm Rads, CPU XSPC RayStorm
Memory 16GB G.Skill RipJaws DDR4-3200
Video Card(s) Sapphire R9 Radeon Vega 64
Storage Samsung SSD 960 EVO NVMe 250GB, Samsung SSD 840 EVO 1TB, Samsung SSD 840 Pro 256GB,4x3TB Seagate
Display(s) BenQ ZOWIE XL2730 144Hz 27 inch Freesync
Case Cooler Master Storm Trooper with side window
Audio Device(s) Turtle Beach Elite Pro Tournament + Elite Pro TAC
Power Supply be quiet! Dark Power Pro 10 1200W
Mouse Logitech G900 Chaos Spectrum
Keyboard ROCCAT Ryos MK Pro
Software Win 10 Pro
#81
Similar ideas to me but im cash strapped so the waterblocks on hold , please consider joining the vega club in the club section , its Getting lonely in there.
I agree that custom water cooling setup is an expensive hobby if one has to buy everything at the same time includig waterblocks, pump, reservoir, radiators, fans etc.
I had all these stuff from before as I am water cooling my PC since some years now.
The only thing I needed was the waterblock for the Vega 64.
Besides current minning craze has driven the prices of used cards very high.
I was able to sell my two year old Fury X for 340€ on eBay.
The whole Vega setup including waterblock cost me at the end only 427€ instead of 767€. :p
 
Joined
Mar 10, 2010
Messages
5,371 (1.78/day)
Likes
1,848
Location
Manchester uk
System Name Quad GT evo V
Processor FX8350 @ 4.6ghz1.435c NB2.44ghz Ht2.64ghz
Motherboard Gigabyte 990X Gaming
Cooling 360EK extreme 360Tt rad all push/pull, cpu,NB/Vrm blocks all EK
Memory Corsair vengeance 16Gb @1600 cas8
Video Card(s) Rx vega 64 waterblockedEK
Storage samsung 840(250OS), WD 1Tb+2Tb +3Tbgrn, 1tb seagate sshd hybrid(games)
Display(s) Samsung uea28"850R 4k freesync, LG 49" 4K 60hz ,Oculus
Case Custom(modded) thermaltake Kandalf
Audio Device(s) Xfi creative 7.1 on board ,Yamaha dts av setup
Power Supply corsair 1200Hxi
Mouse CM optane
Keyboard CM optane
Software Win 10 Pro
Benchmark Scores 15.69K best overall sandra so far 6600 3dmark
#82
I agree that custom water cooling setup is an expensive hobby if one has to buy everything at the same time includig waterblocks, pump, reservoir, radiators, fans etc.
I had all these stuff from before as I am water cooling my PC since some years now.
The only thing I needed was the waterblock for the Vega 64.
Besides current minning craze has driven the prices of used cards very high.
I was able to sell my two year old Fury X for 340€ on eBay.
The whole Vega setup including waterblock cost me at the end only 427€ instead of 767€. :p
yeah thats why i said im in the same boat but i havent got the money just for the block , everything else is already in place and mine sort of cost me too much tbh.
 
Joined
Sep 17, 2014
Messages
5,042 (3.67/day)
Likes
4,041
Location
Duiven, Netherlands
Processor i7 8700k 4.8Ghz @ 1.31v
Motherboard AsRock Fatal1ty K6 Z370
Cooling beQuiet! Dark Rock Pro 3
Memory 16GB Corsair Vengeance LPX 3200/C16
Video Card(s) MSI GTX 1080 Gaming X @ 2100/5500
Storage Samsung 850 EVO 1TB + Samsung 830 256GB + Crucial BX100 250GB + Toshiba 1TB HDD
Display(s) Eizo Foris FG2421
Case Fractal Design Define C TG
Power Supply EVGA G2 750w
Mouse Logitech G502 Protheus Spectrum
Keyboard Sharkoon MK80 (Brown)
Software W10 x64
#83
I agree that custom water cooling setup is an expensive hobby if one has to buy everything at the same time includig waterblocks, pump, reservoir, radiators, fans etc.
I had all these stuff from before as I am water cooling my PC since some years now.
The only thing I needed was the waterblock for the Vega 64.
Besides current minning craze has driven the prices of used cards very high.
I was able to sell my two year old Fury X for 340€ on eBay.
The whole Vega setup including waterblock cost me at the end only 427€ instead of 767€. :p
That's cool math, so your Fury X was free then?

Also, you need to realize that you stepped down a GPU tier; from 980ti equivalent to 1080 (best case!) equivalent. And in that case, 427 eur is not really such a great deal at all. I went from a 780ti (sold for 160 eur) to a 1080 and paid 420 too - also a tier down, but also a full gen older. By comparison your watercooling setup cost you a full GPU gen. This is a performance gain difference of +30-40% (Fury X > Vega) versus +50-60%. I understand its fun to put a block on it, but don't tell anyone its cost effective. You'll never get another 20% OC out of it.
 
Last edited:
Joined
Apr 30, 2011
Messages
940 (0.36/day)
Likes
947
Location
Greece
Processor AMD FX-8350 4GHz@1.3V
Motherboard Gigabyte GA-970A UD3 Rev3.0
Cooling Zalman CNPS5X Performa
Memory 2*4GB Patriot Venom RED DDR3 1600MHz CL9
Video Card(s) XFX RX580 GTS 4GB
Storage Sandisk SSD 120GB, 2 Samsung F1 & F3 (1TB)
Display(s) LG IPS235
Case Zalman Neo Z9 Black
Audio Device(s) Via 7.1 onboard
Power Supply OCZ Z550
Mouse Zalman ZM-M401R
Keyboard Trust GXT280
Software Win 7 sp1 64bit
Benchmark Scores CB R15 64bit: single core 99p, multicore 626p WPrime 1.55 (8 cores): 9.0 secs
#84
Intel and NVIDIA also have brilliant ideas. It's just that their products work, so we're not talking about "great ideas" so much. With AMD you're forced to droll about all these technologies inside, because the sum of them, called "Vega", just doesn't stand out, so no one would care.

To succeed in the compute segment they'll need a card made for compute segment. RX Vega is not that, AMD has pro cards in the lineup.

LOL. You've just said it. Some dies will be stable at lower voltage, some won't. That's true for all GPUs and all CPUs.
Yet, you and @HD64G are trying to make a general rule out of it.
You can buy your dreamed Vega and do with whatever you want.
Companies buying Vega-based cards for computation will not undervolt or overclock them. For them only the stock performance and efficiency matter.
Companies will buy the very cheap -for what they will use it for- Vega FE and will be glad for that without any need to tinker it. PC enthusiasts on the other side use to take advantage of any strong aspect of the product they buy to extract maximum performance while maintaining its reliability and not destroying its longetivity. Vega RX are having more potential than what is shown in reviews. Tbh, our @W1zzard proved that even without any tinkering just by testing all settings on both bios. In conclusion, Vega RX are for those guys not in need of 1080Ti performance levels (4K or 144Hz screen) who are happy to play with it to make it much better than in default state instead of plug&play. I say that again: Reminds me of 7950-7970 launch, where they were bashed for being hot and power hungry and equal to 670-680, but after 6 months or so, they were the best for any pc ehtusiast and nVidia owners had only one argument: drivers.
 
Joined
Jan 6, 2014
Messages
73 (0.04/day)
Likes
62
Location
Germany
System Name Main Machine
Processor Intel i7-7700K
Motherboard ASUS ROG MAXIMUS IX CODE
Cooling Water cooling, EK-DDC 3.2 PWM, 2x240mm+1x120mm Rads, CPU XSPC RayStorm
Memory 16GB G.Skill RipJaws DDR4-3200
Video Card(s) Sapphire R9 Radeon Vega 64
Storage Samsung SSD 960 EVO NVMe 250GB, Samsung SSD 840 EVO 1TB, Samsung SSD 840 Pro 256GB,4x3TB Seagate
Display(s) BenQ ZOWIE XL2730 144Hz 27 inch Freesync
Case Cooler Master Storm Trooper with side window
Audio Device(s) Turtle Beach Elite Pro Tournament + Elite Pro TAC
Power Supply be quiet! Dark Power Pro 10 1200W
Mouse Logitech G900 Chaos Spectrum
Keyboard ROCCAT Ryos MK Pro
Software Win 10 Pro
#85
That's cool math, so your Fury X was free then?
The Fury X at that time cost me around 600€, but in normal market conditions I would not expect to get 340€ back specially as the card is more than two years old.
That is more than 50% return of money. I don't think this was possible without GPU prices being that high at the moment.

I usually sell my old high end GPUs on eBay and usually don't get back so much money if they are older than two years and no more warranty.
In Germany we have two years warranty.
 
Joined
Sep 17, 2014
Messages
5,042 (3.67/day)
Likes
4,041
Location
Duiven, Netherlands
Processor i7 8700k 4.8Ghz @ 1.31v
Motherboard AsRock Fatal1ty K6 Z370
Cooling beQuiet! Dark Rock Pro 3
Memory 16GB Corsair Vengeance LPX 3200/C16
Video Card(s) MSI GTX 1080 Gaming X @ 2100/5500
Storage Samsung 850 EVO 1TB + Samsung 830 256GB + Crucial BX100 250GB + Toshiba 1TB HDD
Display(s) Eizo Foris FG2421
Case Fractal Design Define C TG
Power Supply EVGA G2 750w
Mouse Logitech G502 Protheus Spectrum
Keyboard Sharkoon MK80 (Brown)
Software W10 x64
#86
I say that again: Reminds me of 7950-7970 launch, where they were bashed for being hot and power hungry and equal to 670-680, but after 6 months or so, they were the best for any pc ehtusiast and nVidia owners had only one argument: drivers.
Except Nvidia didn't refresh the 670 and 680 two or three times did they. Because it was only the Ghz editions of the 7970 that would win the day, and yes, agreed, the 7950 was just a great value high end card because it sported almost the same core. And then, not too long after that, the 770 crushed them all at a super low price, while the 780ti cannibalized Hawaii. Now go put a 670/680 side by side today with a 7970 and watch them perform in the exact same ballpark - still. But the 7970 does want more juice. In the end its real life advantage was extremely short lived. Kepler Refresh was the beginning of the huge gap that is now between AMD and NV. You can completely forget about a 7970 happening again with Vega, and I really hope that wasn't in your mind as a sane thought. Since Kepler Refresh the clockspeed gap has been quadrupled, from 100 to 400mhz, and during Maxwell the gap even ran up to 500mhz versus your average 980ti OC.

The Fury X at that time cost me around 600€, but in normal market conditions I would not expect to get 340€ back expecially as the card is more that two years old.
That is more than 50% return of money. I don't think this was possible without GPU prices being that high at the moment.
Yeah you're right about that, I just fell over the comment about water coming out cheap for you, because it really doesn't :) Its easy to fool ourselves like that.
 
Last edited:
Joined
Jan 8, 2017
Messages
2,464 (4.66/day)
Likes
1,661
System Name Good enough
Processor AMD FX-6300 - 4.5 Ghz
Motherboard ASRock 970M Pro3
Cooling Scythe Katana 4 - 3x 120mm case fans
Memory 16GB - 4x4GB A-DATA 1866 Mhz (OC)
Video Card(s) ASUS GTX 1060 Turbo 6GB ~ 2139 Mhz / 9.4 Gbps
Storage 1x Samsung 850 EVO 250GB , 1x 1 Tb Seagate something or other
Display(s) 1080p TV
Case Zalman R1
Power Supply 500W
#87
Pretty sure 780ti didn't cannibalized anything , 290X was cheaper and it was neck and neck in terms of performance in many cases.
 
Last edited:
Joined
Jun 28, 2016
Messages
1,209 (1.67/day)
Likes
352
Location
Warsaw, Poland
Processor i5-7500
Motherboard ASRock H270M-ITX/ac
Cooling stock
Memory 16GB
Video Card(s) MSI GTX 1050 Gaming
Storage a lot...
Display(s) EIZO EV2023W
Case Fractal Design Node 304
Power Supply be quiet! Pure Power 10 400W
Keyboard Logitech UltraX Flat
Benchmark Scores good enough
#88
Companies will buy the very cheap -for what they will use it for- Vega FE and will be glad for that without any need to tinker it. PC enthusiasts on the other side use to take advantage of any strong aspect of the product they buy to extract maximum performance while maintaining its reliability and not destroying its longetivity. Vega RX are having more potential than what is shown in reviews.
OC enthusiasts, not PC enthusiasts...

And some very bad news for you: companies won't buy Vega FE. This card is only a showcase of available technology.
You must have missed the launch of workstation Vega: FirePro WX9100. It's already here and it won't be cheap (although possibly cheaper than competing Quadro - as usual).
 
Joined
Jan 6, 2014
Messages
73 (0.04/day)
Likes
62
Location
Germany
System Name Main Machine
Processor Intel i7-7700K
Motherboard ASUS ROG MAXIMUS IX CODE
Cooling Water cooling, EK-DDC 3.2 PWM, 2x240mm+1x120mm Rads, CPU XSPC RayStorm
Memory 16GB G.Skill RipJaws DDR4-3200
Video Card(s) Sapphire R9 Radeon Vega 64
Storage Samsung SSD 960 EVO NVMe 250GB, Samsung SSD 840 EVO 1TB, Samsung SSD 840 Pro 256GB,4x3TB Seagate
Display(s) BenQ ZOWIE XL2730 144Hz 27 inch Freesync
Case Cooler Master Storm Trooper with side window
Audio Device(s) Turtle Beach Elite Pro Tournament + Elite Pro TAC
Power Supply be quiet! Dark Power Pro 10 1200W
Mouse Logitech G900 Chaos Spectrum
Keyboard ROCCAT Ryos MK Pro
Software Win 10 Pro
#89
Yeah you're right about that, I just fell over the comment about water coming out cheap for you, because it really doesn't :) Its easy to fool ourselves like that.
You are right water never comes cheap, but all the money that you get back from selling old stuff could be part of the investment in the new stuff.
Every penny that I can get back from selling older GPU is a welcome rebate for the new one. :)
 
Joined
Sep 17, 2014
Messages
5,042 (3.67/day)
Likes
4,041
Location
Duiven, Netherlands
Processor i7 8700k 4.8Ghz @ 1.31v
Motherboard AsRock Fatal1ty K6 Z370
Cooling beQuiet! Dark Rock Pro 3
Memory 16GB Corsair Vengeance LPX 3200/C16
Video Card(s) MSI GTX 1080 Gaming X @ 2100/5500
Storage Samsung 850 EVO 1TB + Samsung 830 256GB + Crucial BX100 250GB + Toshiba 1TB HDD
Display(s) Eizo Foris FG2421
Case Fractal Design Define C TG
Power Supply EVGA G2 750w
Mouse Logitech G502 Protheus Spectrum
Keyboard Sharkoon MK80 (Brown)
Software W10 x64
#90
Pretty sure 780ti didn't cannibalized anything , 290X was cheaper and it was neck and neck in terms of performance in many cases.
Almost nobody bought a 290X, only 290s, and both were way too hot. A vast number of these cards, along with 280x's was used to mine Bitcoin. Meanwhile, Nvidia was spinning the 780 and the 780ti as the poor man's Titans that did better than the 1K dollar original in gaming. You're right tho, it was the 780 that did the work, not so much the ti. Also, Kepler Refresh released earlier than Hawaii did.

And even in the midst of the Bitcoin craze, look at that market share go. It managed to stabilize for 2 quarters in 2013/14 instead of continuing the steady drop... , now, consider that mining was impossible on Nvidia altogether. AMD's GCN has only really thrived shortly after the 7950 was available, until the 7970 Ghz. After that, loss upon loss.

 
Joined
Jan 8, 2017
Messages
2,464 (4.66/day)
Likes
1,661
System Name Good enough
Processor AMD FX-6300 - 4.5 Ghz
Motherboard ASRock 970M Pro3
Cooling Scythe Katana 4 - 3x 120mm case fans
Memory 16GB - 4x4GB A-DATA 1866 Mhz (OC)
Video Card(s) ASUS GTX 1060 Turbo 6GB ~ 2139 Mhz / 9.4 Gbps
Storage 1x Samsung 850 EVO 250GB , 1x 1 Tb Seagate something or other
Display(s) 1080p TV
Case Zalman R1
Power Supply 500W
#91
Almost nobody bought a 290X, only 290s, and both were way too hot. A vast number of these cards, along with 280x's was used to mine Bitcoin. Meanwhile, Nvidia was spinning the 780 and the 780ti as the poor man's Titans that did better than the 1K dollar original in gaming. You're right tho, it was the 780 that did the work, not so much the ti. Also, Kepler Refresh released earlier than Hawaii did.

And even in the midst of the Bitcoin craze, look at that market share go. It managed to stabilize for 2 quarters in 2013/14 instead of continuing the steady drop... , now, consider that mining was impossible on Nvidia altogether. AMD's GCN has only really thrived shortly after the 7950 was available, until the 7970 Ghz. After that, loss upon loss.

I wasn't really speaking about market share , just saying the 780ti didn't cannibalize Hawaii based graphics cards. 780ti might had a slight edge , but came at a steep price. At 700$ launch price I wouldn't call it the poor man's Titan , I think the 290X which came at almost half the price of a Titan was the true poor man's Titan. They were hot , but when you put it into perspective in terms of performance and price , they were pretty damn good. I have seen more people using the 290/290X rather than the 780/780ti.
 
Joined
Sep 17, 2014
Messages
5,042 (3.67/day)
Likes
4,041
Location
Duiven, Netherlands
Processor i7 8700k 4.8Ghz @ 1.31v
Motherboard AsRock Fatal1ty K6 Z370
Cooling beQuiet! Dark Rock Pro 3
Memory 16GB Corsair Vengeance LPX 3200/C16
Video Card(s) MSI GTX 1080 Gaming X @ 2100/5500
Storage Samsung 850 EVO 1TB + Samsung 830 256GB + Crucial BX100 250GB + Toshiba 1TB HDD
Display(s) Eizo Foris FG2421
Case Fractal Design Define C TG
Power Supply EVGA G2 750w
Mouse Logitech G502 Protheus Spectrum
Keyboard Sharkoon MK80 (Brown)
Software W10 x64
#92
I wasn't really speaking about market share , just saying the 780ti didn't cannibalize Hawaii based graphics cards. 780ti might had a slight edge , but came at a steep price. At 700$ launch price I wouldn't call it he poor man's Titans , I think the 290X which came at almost half the price of a Titan was the true poor man's Titan. They were hot , but when you put it into perspective in terms of performance and price , they were pretty damn good. I have seen more people using the 290/290X rather than 780/780ti.
I've seen the ratio of second hand sales of 290/290X in the months following its release, compared to the second hand sales of 780/780ti's. It was 9 to 1 in the Netherlands for over 6 months. Nine. To. One. That's how great they were. There was a good reason they were cheap.
 
Joined
Jan 8, 2017
Messages
2,464 (4.66/day)
Likes
1,661
System Name Good enough
Processor AMD FX-6300 - 4.5 Ghz
Motherboard ASRock 970M Pro3
Cooling Scythe Katana 4 - 3x 120mm case fans
Memory 16GB - 4x4GB A-DATA 1866 Mhz (OC)
Video Card(s) ASUS GTX 1060 Turbo 6GB ~ 2139 Mhz / 9.4 Gbps
Storage 1x Samsung 850 EVO 250GB , 1x 1 Tb Seagate something or other
Display(s) 1080p TV
Case Zalman R1
Power Supply 500W
#93
I've seen the ratio of second hand sales of 290/290X in the months following its release, compared to the second hand sales of 780/780ti's. It was 9 to 1 in the Netherlands for over 6 months. Nine. To. One. That's how great they were.
Not sure what you are trying to say. Of course there were way more second hand 290/290X's than 780/780ti's , you said it yourself they were bought like crazy by miners. That ratio has nothing to do with how good they were.

There was a good reason they were cheap.
They were cheap compared to the competition even if you look at launch prices.

My point was that they were fast and cheap. Saying that a 780ti cannibalized them would be grossly exaggerated . For that to be the case it would have had to be faster and cheaper , that was not the case.
 
Last edited:
Joined
Sep 17, 2014
Messages
5,042 (3.67/day)
Likes
4,041
Location
Duiven, Netherlands
Processor i7 8700k 4.8Ghz @ 1.31v
Motherboard AsRock Fatal1ty K6 Z370
Cooling beQuiet! Dark Rock Pro 3
Memory 16GB Corsair Vengeance LPX 3200/C16
Video Card(s) MSI GTX 1080 Gaming X @ 2100/5500
Storage Samsung 850 EVO 1TB + Samsung 830 256GB + Crucial BX100 250GB + Toshiba 1TB HDD
Display(s) Eizo Foris FG2421
Case Fractal Design Define C TG
Power Supply EVGA G2 750w
Mouse Logitech G502 Protheus Spectrum
Keyboard Sharkoon MK80 (Brown)
Software W10 x64
#94
Not sure what you are trying to say. Of course there were way more second hand 290/290X's than 780/780ti's , you said it yourself they were bought like crazy by miners. That ratio has nothing to do with how good they were.

My point was that they were fast and cheap , saying that a 780ti cannibalized them would be grossly exaggerated.
I'll try again. Even with 9 times more Hawaiis on second hand getting sold off, AMD's market share hasn't seen a single % increase after its release. Meanwhile, at Nvidia's release of 770/780, they gained share at the cost of AMD. So those were either ALL 770s they sold, or it was something else... I understand your point about Hawaii's price/perf metric, but it didn't result in any market share movement EVEN while the vast majority of these cards were used to mine.

The last surge in AMD's share was by releasing the 280 and the refresh + price drop. AMD had a better midrange, but their high end did not sell to gamers at all since the post 7970-era.
 
Joined
Jan 8, 2017
Messages
2,464 (4.66/day)
Likes
1,661
System Name Good enough
Processor AMD FX-6300 - 4.5 Ghz
Motherboard ASRock 970M Pro3
Cooling Scythe Katana 4 - 3x 120mm case fans
Memory 16GB - 4x4GB A-DATA 1866 Mhz (OC)
Video Card(s) ASUS GTX 1060 Turbo 6GB ~ 2139 Mhz / 9.4 Gbps
Storage 1x Samsung 850 EVO 250GB , 1x 1 Tb Seagate something or other
Display(s) 1080p TV
Case Zalman R1
Power Supply 500W
#95
I'll try again. Even with 9 times more Hawaiis on second hand getting sold off, AMD's market share hasn't seen a single % increase after its release. Meanwhile, at Nvidia's release of 770/780, they gained share at the cost of AMD. So those were either ALL 770s they sold, or it was something else... I understand your point about Hawaii's price/perf metric, but it didn't result in any market share movement EVEN while the vast majority of these cards were used to mine.
I am not arguing with you about the market share , you have to take into consideration that that was also about the time when AMD had stopped pushing their mobile GPUs , I am sure a good chunk of that market share from then on came from laptops in Nvidia's case.

Thing is 780/780tis are a rare sight , I've just never seen many around even when they were the new hot thing , they're also quite scarce on the second hand market just as you said , that tells me they didn't sell many of those.

If you're saying that Nvidia's flagships were more successful from a business point of view , yeah they might have been but as actual products I think AMD had some more than solid offerings in that period too.
 
Last edited:
Joined
Aug 6, 2009
Messages
677 (0.21/day)
Likes
154
Location
Chicago, Illinois
#96
I agree that custom water cooling setup is an expensive hobby if one has to buy everything at the same time includig waterblocks, pump, reservoir, radiators, fans etc.
I had all these stuff from before as I am water cooling my PC since some years now.
The only thing I needed was the waterblock for the Vega 64.
Besides current minning craze has driven the prices of used cards very high.
I was able to sell my two year old Fury X for 340€ on eBay.
The whole Vega setup including waterblock cost me at the end only 427€ instead of 767€. :p
Ha,some clowns over on Linus Tech were trying to argue that for the price of the Monsta 420 AIO you could build a custom loop.
 
Joined
Aug 22, 2015
Messages
2,815 (2.72/day)
Likes
1,497
Location
Europe
System Name eazen corp | Xentronon 6.7
Processor Intel Core i7 3960X, overclocked at 4.5 GHz
Motherboard ASUS P9X79
Cooling Noctua NH-D14 SE2011 / 3x Corsair AF140L case fans (2 blowing in, 1 out)
Memory G.Skill Sniper 4x4 GB DDR3 1866, CL9-10-9-9-2T, 1.5 V
Video Card(s) G1 Gaming GeForce GTX 980 Ti, overclocked to ~1350-1400 MHz Core / 1928 MHz RAM for daily usage
Storage Samsung SSD 830, 128GB & SSD 850 Evo, 500GB & F3, 1TB 7200RPM / Seagate Barracuda Green, 2TB 5900RPM
Display(s) iiyama 27" 1080p 120Hz & 22" 1680x1050 60Hz / Samsung 40" 1080p HDTV
Case Corsair Carbide 600C, inverted design case
Audio Device(s) Creative SB X-Fi Xtreme Music/Sony HD AVR@PC&HDTV @ Teufel Concept F &Theater 80 & Razer Carcharias
Power Supply EVGA 650 GQ in semi-passive mode
Mouse Logitech G700 @ Steelseries DeX
Keyboard Logitech G11
Software Microsoft Windows 10 Pro
Benchmark Scores 4144 UG Superposition 1080p Extreme|6302 Time Spy|P16615 Firestrike|P21524 3DM11|X37002 3DM Vantage
#97
Companies will buy the very cheap -for what they will use it for- Vega FE and will be glad for that without any need to tinker it. PC enthusiasts on the other side use to take advantage of any strong aspect of the product they buy to extract maximum performance while maintaining its reliability and not destroying its longetivity. Vega RX are having more potential than what is shown in reviews. Tbh, our @W1zzard proved that even without any tinkering just by testing all settings on both bios. In conclusion, Vega RX are for those guys not in need of 1080Ti performance levels (4K or 144Hz screen) who are happy to play with it to make it much better than in default state instead of plug&play. I say that again: Reminds me of 7950-7970 launch, where they were bashed for being hot and power hungry and equal to 670-680, but after 6 months or so, they were the best for any pc ehtusiast and nVidia owners had only one argument: drivers.
And to add on top of that: HD 7970 was the first 28nm GPU, the first GPU to break 1000 MHz and it was the GPU king when launched, easily beating the GTX 580. GTX 680 came later to TIE the 7970, not really taking the crown back. Especially smart users didn't buy the GTX 680, because it had 1 GB less vram and a smaller bus (256 vs 384 bit). I knew back then that the HD 7970 would be the easily smarter buy over time, and history proves me right. Actually, this is the last big hit of AMD, I wouldn't count R9 290X into it, despite it being on top of GTX 780 (Titan is irrelevant because of too high pricing I'd say) - it was simply too hot and too loud. Anandtech said the same - there's a limit that can be breached, when a card, despite its high performance gets unwanted by being too loud and too hot. IMO, 290s were powerhouses, as well as 390s, but first of them were too loud, whereas later cards too inefficient, especially comparing Multi Monitor / Video usage to Nvidia cards of 700 and 900 series.

@Vayra86 is right when he says that GTX 700 gen dominated the market and won against the 200 series. It's a fact by the numbers, no need to have a big argument about it. AMD lost a lot of market share between release of HD 7970 and R9 200 series, and it went even more downhill after that.

One more thing: Vega is a very nice and very efficient GPU - if you downvolt it. The primitive "GPU Boost" (I like to call it "GPU Boost 0.5") of AMD can't do it, whereas Nvidia cards always run on lowest voltage possible with highest clocks possible, making them super efficient (that is true since 600 series, and no wonder, they won every efficiency benchmark since then). As a matter of fact, this is part of Nvidias big success. Would AMD have the same architecture they now have, and the same GPU Boost of Nvidia, their cards would be FAR FAR better. People don't seem to recognize this, but AMD software is basically what sucks, not their hardware, not at all. Vega is a very strong GPU, as was Fiji, Hawaii, Tahiti - no matter what you name, all these architectures are powerhouses (both in power consumption AND performance). ;)
 
Joined
Dec 22, 2011
Messages
2,295 (0.97/day)
Likes
1,313
System Name Zimmer Frame Rates
Processor Intel i7 920 @ Stock speeds baby
Motherboard EVGA X58 3X SLI
Cooling True 120
Memory Corsair Vengeance 12GB
Video Card(s) Palit GTX 980 Ti Super JetStream
Storage Of course
Display(s) Crossover 27Q 27" 2560x1440
Case Antec 1200
Audio Device(s) Don't be silly
Power Supply XFX 650W Core
Mouse Razer Deathadder Chroma
Keyboard Logitech UltraX
Software Windows 10
Benchmark Scores Epic
#98
These history lessons are great, it's almost like after the Fermi debacle Nvidia learnt a lesson and decided to differentiate their prosumer oriented cards with the consumers oriented cards.

Now their middle of range consumer focused GX104 chips have been trading blows with top of the range AMD GPUs ever since, who'd a thunk it.
 
Last edited:
Joined
Aug 22, 2015
Messages
2,815 (2.72/day)
Likes
1,497
Location
Europe
System Name eazen corp | Xentronon 6.7
Processor Intel Core i7 3960X, overclocked at 4.5 GHz
Motherboard ASUS P9X79
Cooling Noctua NH-D14 SE2011 / 3x Corsair AF140L case fans (2 blowing in, 1 out)
Memory G.Skill Sniper 4x4 GB DDR3 1866, CL9-10-9-9-2T, 1.5 V
Video Card(s) G1 Gaming GeForce GTX 980 Ti, overclocked to ~1350-1400 MHz Core / 1928 MHz RAM for daily usage
Storage Samsung SSD 830, 128GB & SSD 850 Evo, 500GB & F3, 1TB 7200RPM / Seagate Barracuda Green, 2TB 5900RPM
Display(s) iiyama 27" 1080p 120Hz & 22" 1680x1050 60Hz / Samsung 40" 1080p HDTV
Case Corsair Carbide 600C, inverted design case
Audio Device(s) Creative SB X-Fi Xtreme Music/Sony HD AVR@PC&HDTV @ Teufel Concept F &Theater 80 & Razer Carcharias
Power Supply EVGA 650 GQ in semi-passive mode
Mouse Logitech G700 @ Steelseries DeX
Keyboard Logitech G11
Software Microsoft Windows 10 Pro
Benchmark Scores 4144 UG Superposition 1080p Extreme|6302 Time Spy|P16615 Firestrike|P21524 3DM11|X37002 3DM Vantage
#99
These history lessons are great, it's almost like after the Fermi debacle Nvidia learnt a lesson and decided to differentiate their prosumer oriented cards with the consumers oriented cards.
They released GPU Boost with 600 series. That's what set them apart from GPUs like HD 7970 who simply tried to run on highest clocks always, nevermind if its efficient or not. Since then they ruled - and AMD to this day still hasn't copied the feature, just half of it. On top of that, running a GPU on lowest voltage possible also gives you higher clocks, *in average*. Something every good AMD user knows, but they all have to do it manually. Right from the start, AMD cards aren't tuned to be at maximum efficiency, they are just simply tuned to "function". Nvidia cards were like that prior to GTX 600 series as well.
 
Joined
Dec 22, 2011
Messages
2,295 (0.97/day)
Likes
1,313
System Name Zimmer Frame Rates
Processor Intel i7 920 @ Stock speeds baby
Motherboard EVGA X58 3X SLI
Cooling True 120
Memory Corsair Vengeance 12GB
Video Card(s) Palit GTX 980 Ti Super JetStream
Storage Of course
Display(s) Crossover 27Q 27" 2560x1440
Case Antec 1200
Audio Device(s) Don't be silly
Power Supply XFX 650W Core
Mouse Razer Deathadder Chroma
Keyboard Logitech UltraX
Software Windows 10
Benchmark Scores Epic
And that's all just swell, but it doesn't change what I said.
 
Top