Another GK106 Completes the Stack
It has been an interesting year for graphics cards and 2012 still has another solid quarter of releases ahead of it. With the launch of AMD's 7000-series back in January, followed by the start of NVIDIA's Kepler lineup in March, we have had new graphics cards on a very regular basis ever since. And while AMD's Radeon HD 7000 cards seemed to be bunched together a bit better, NVIDIA has staggered the release of the various Kepler cards, either because of capacity at the manufacturing facilities or due to product marketing plans - take your pick.
Today we see the completion of the NVIDIA GeForce GTX 600 stack (if you believe the PR at NVIDIA) with the release of the GeForce GTX 650 Ti, a $150 graphics card that fills in the gap between the somewhat anemic GTX 650 and GT 640 cards and the most recently unveiled card, the GTX 660 2GB that currently sells for $229.
The GTX 650 Ti has more in common with the GTX 660 than it does the GTX 650, both being based on the GK106 GPU, but is missing some of the unique features that NVIDIA has touted of the 600-series cards like GPU Boost. Let's dive into the product and see if this new card will be the best option for those of you with $150 graphics budgets.
Subject: Graphics Cards | October 7, 2012 - 10:37 PM | Tim Verry
Tagged: nvidia, kepler, gtx 650ti, gpu, gk106-220
The NVIDIA GeForce GTX 650 Ti is rumored to launch soon, and so far specifications have leaked on the reference design as well as two custom cards from ASUS and Galaxy. Zotac is the latest manufacturer to have its GTX 650 Ti lineup leaked, and the company is bringing as many as three graphics cards to the GK106-220 Kepler family. In all, Zotac is rumored to be launching one 1GB GTX 650 Ti and two 2GB GPUs – all with vared levels of factory overclocks. Video outputs on all three cards include two DVI and two HDMI connectors.
The Zotac GTX 650 Ti 1GB stays close to the reference design, but bumps up the GPU core clockspeed to 941 MHz. It also includes 1 GB of GDDR5 memory on a 128-bit interface clocked at 1350 MHz (5400 MHz effective), which matches the reference design. The price of this card is said to be $160, and features a custom cooler from Zotac that is similar (but smaller than) to the cooler used on the company's GTX 660 Ti GPU wich we recently reviewed.
The Zotac GTX 650 Ti 2GB is, as the name suggests, a GTX 650 Ti graphics card with 2GB of GDDR5 memory. It features Zotac's custom cooler, and a single PCI-E 6-pin power connector. The GPU clockspeed is 941 MHz and the memory clockspeed is 1350 MHz. The extra 1GB of graphics memory is nice, but it is still on a 128-bit interface so don't expect too much of a performance boost. MSRP of this card is rumored to be $180.
Finally, the GTX 650 Ti 2GB AMP! Edition is Zotac's highest-end GTX 650 Ti graphics card. It comes with the GK106-220 Kepler GPU and 2GB of GDDR5 memory on a 128-bit bus. Powered by a single 6-pin PEG connector, the factory overclocked graphics card is clocked at 1033 MHz for the GPU and 1550 MHz (6200 MHz effective) for the memory.The Zotac GTX 650 Ti AMP! Edition comes with the company's custom cooler and is the first card to feature factory overclocked memory. The rumored price of this card is $190. Unfortunately, that puts it fairly close to the price of a reference GTX 660, which may make this card a hard sell. The factory overclocks are impressive, but saving up the extra $30 needed to get a GTX 660 is likely a better idea because it will still offer up better performance thanks to the additional CUDA cores and wider memory bus.
The following chart compares the three Zotac cards to the leaked reference specifications.
|Reference Specifications||Zotac GTX 650 Ti 1GB||Zotac GTX 650 Ti 2GB||Zotac GTX 650 Ti 2GB AMP! Edition|
|CPU Clockspeed||925 MHz||941 MHz||941 MHz||1033 MHz|
|Memory Clockspeed||1350 MHz||1350 MHz||1350 MHz||1550 MHz|
|GDDR5 Amount||1 GB||1 GB||2 GB||2 GB|
Comparison of several GTX 650 Ti graphics cards versus the rumored reference specifications.
Further, this chart compares the leaked specifications of the top end cards from each manufacturer (at least, the ones we know of so far) to the highest-end Zotac GPU: the 2GB AMP! Edition.
|Reference Specifications||ASUS GTX 650 Ti TOP||Galaxy GTX 650 Ti GC 1GB||Gigabyte GTX 650 Ti OC||Zotac GTX 650 Ti 2GB AMP! Edition||POV GTX 650 Ti 1GB Ultra Charged|
|CPU Clockspeed||925 MHz||1033 MHz||966 MHz||1032 MHz||1033 MHz||1058 MHz|
|Memory Clockspeed||1350 MHz||1350 MHz||1350 MHz||1350 MHz||1550 MHz||1350 MHz|
|GDDR5 Amount||1 GB||1 GB||1 GB||2 GB||2 GB||1 GB|
|Video Outputs||2 x DVI, 1 x HDMI||2 x DVI, 1 x HDMI, 1 x VGA||2 x DVI, 1 x HDMI||2 x DVI, 1 x HDMI, 1 x VGA||2 x DVI, 2 x HDMI||1 x DVI, 1 x HDMI, 1 x VGA|
Inno3D is also rumored to have a GTX 650 Ti graphics card coming out, but we don't know clockspeeds or price on it. Only that it has two DVI and one HDMI connector, a single PEG power connector, and a custom cooler.
Overall, the Zotac card measures up well, with pricing being the only major disadvantage. The 2GB of memory, factory overclocks, and two HDMI ports are welcome additions, however. Interestingly, the Zotac card is not the highest clocked graphics card overall, but it is the only one that features overclocked memory. It is unclear to me why manufactuers of NVIDIA cards are so hesitant to push the memory clockspeeds (or if they are even allowed to), but Zotac seems to prove that it is possible to do so.
Also worth pointing out is the rumored pricing, as some of these custom graphics cards are pushing $200 (especially the ASUS card when coverted to USD... I'm sure that has to be in error...), and reference GTX 660 with the full GK106 Kepler core are only $230. It will be interesting to see if these rumored prices turn out to be true, and how well Zotac's factory overclocked 650 Ti models sell.
Subject: General Tech | October 4, 2012 - 10:08 PM | Tim Verry
Tagged: nvidia, kepler, gtx 650ti, gpu, geforce
Earlier this year, specifications for an as-yet-unreleased GTX 650 Ti graphics card from NVIDIA leaked. At the time, the rumors indicated that the GTX 650 Ti would have hardware closer to the GTX 650 than the GTX 660 but still be based on the GK106 Kepler chip. It would have a 128-bit memory interface, 48 testure units, and 576 CUDA cores in 1.5 GPCs (3 SMX units). And to top it off, it had a rumored price of around $170! Not exactly a bargain.
Welll, as the launch gets closer more details are being leaked, and this time around the rumored information is indicating that the GTX 650 Ti will be closer in performance to the GTX 660 and cost around $140-$150. That certainly sounds better!
The new rumors are indicating that the reference GTX 650 Ti will have 768 CUDA cores, and 64 texture units, which means it has the full two GPCs (so it is only missing the one-half of a GPC that you get with GTX 660). and four SMX units. As a point of reference, the GTX 660 – which NVIDIA swears is the full GK106 chip – has five SMX units in 2 and a half GPCs.
The following image shows the layout of the GTX 660. The GTX 650 Ti will have the GPC on the far right disabled. Previous rumors suggested that the entire middle GPC would be turned off, so the new rumors are definitely looking more promising in terms of potential performance.
Specifically marked GK106-220 on the die, the GTX 650 Ti is based the same GK106 Kepler chip as the GTX 660, but with some features disabled. The GPU is reportedly clocked at 925MHz, and it does not support NVIDIA's GPU Boost technology.
Memory performance will take a large hit compared to the full GK106 chip. The GTX 650 Ti will feature 1GB of GDDR5 memory clocked at 1350MHz on a 128-bit memory interface. That amounts to approximately 86.4 GB/s bandwidth, which is slightly over half of the GTX 660's 144.2 GB/s bandwidth. Also, it's just barely over the 80 GB/s bandwidth of the GTX 650 (which makes sense, considering they are both using 128-bit interfaces).
The latest rumors indicate the GTX 650 Ti will be priced at around $140 with custom cards such as recently leaked Galaxy GTX 650 Ti GC on Newegg costing more ($149). These new leaked specifications have more weight than the previous rumors since they have come from multiple leaks from multiple places, so I am hoping that these new rumors are the real deal. If so, the GTX 650 Ti becomes a much better value that it was rumored to be before!
You can find more photos of a leaked GTX 650 Ti over at Chiphell.
Subject: Graphics Cards | October 4, 2012 - 04:42 PM | Tim Verry
Tagged: nvidia, kepler, gtx 680, gtx 670, gtx 660 Ti, gigabyte, factory overclocked
Gigabyte is launching three new factory overclocked graphics cards featuring a Kepler GPU, custom PCB, and custom cooler. The factory overclocks are notable, but will cost you. Specifically, the company is producing versions of the GTX 660 Ti, GTX 670, and GTX 680.
The Gigabyte GV-N680OC-4GD takes the GTX 680 GPU, places it on a custom PCB, and pairs it with 4GB of GDDR5 memory. It features two 6-pin PCI-E power connectors, and Gigabyte’s Windforce X3 450W custom cooler using a triangular fin design that allegedly increases cooling potential. While the GDDR5 memory clockspeeds have not been increased over the reference clocks, the GPU core and boost clockspeeds have been pushed to 1071 MHz and 1137 MHz respectively. The following chart shows the differences in clockspeed and memory over the reference design.
|Reference GTX 680||Gigabyte N680OC-4GD|
|GPU Core||1006 MHz||1071 MHz|
|GPU Boost||1058 MHz||1137 MHz|
|GDDR5 Amount||2 GB||4 GB|
|GDDR5 Speed||6 Gbps||6 Gbps|
The GTX 680 is not the only card to get a custom makeover by Gigabyte, however. The GV-N670OC-4GD is a custom GTX 670. With this card, Gigabyte has set the base clockspeed at 980 MHz – the boost clockspeed of reference cards – and the boost clockspeed at 1058 MHz. Gigabyte has also doubled down on the GDDR5 memory by packing 4GB onto the custom PCB. The memory clockspeed remains the same 6 Gbps as reference cards, however.
This card uses the same Windforce X3 cooler as the cust GTX 680, and as a result has a triple slot design that looks identical to the N680OC-4GD. If you look just above the PCI-E connector though, you can see tell them apart by the product name.
|Reference GTX 670||Gigabyte N670OC-4GD|
|GPU Core||915 MHz||980 MHz|
|GPU Boost||980 MHz||1058 MHz|
|GDDR5 Amount||2 GB||4 GB|
|GDDR5 Speed||6 Gbps||6 Gbps|
Finally, we have the GV-N66TOC-3GD which overclocks the GTX 660 Ti GPU to the max. Factory clockspeeds are set at 1032 MHz base and 1111 MHz boost. Memory also sees a small bump from 2GB reference to 3GB. On the other hand, the memory is not overclocked and remains at the reference 6 Gbps clockspeed. This card also has a triple fan Windforce cooler, however this version is not the triple slot design found on the GTX 670 and GTX 680s SKUs – only dual slot.
|Reference GTX 660 Ti||Gigabyte N66TOC-3GD|
|GPU Core||915 MHz||1032 MHz|
|GPU Boost||980 MHz||1111 MHz|
|GDDR5 Amount||2 GB||3 GB|
|GDDR5 Speed||6 Gbps||6 Gbps|
All three of the Gigabyte GPUs feature two DVI, one full-size HDMI, and one full-size DisplayPort connector for video outputs.
All three factory overclocked graphics cards feature respectable GPU overclocks, and it appears that Gigabyte has provided ample cooling for each GPU. The triple slot, triple fan version on the N670OC-4GD and N680OC-4GD in particular seem to offer headroom above even what Gigabyte has clocked these out of the box. Curiously though, Gigabyte is continuing the trend of not touching the memory clockspeed of Kepler cards. It may be that the RAM chips are already at their max on the reference design, or there could be some behind the scenes talk with NVIDIA not waning Add In Board partners to touch the memory Unfortunately, all I have at this point is speculation, but it is a rather curious omission on such high end cards. That point becomes clearer when price is taken into consideration. Videocardz claims to have the pricing information for the three video cards, and the custom cards are going to cost you a large premium over reference cards. The rumored prices can be found in the charts above compared against the reference pricing, but the basic run down is that the GV-N66TOC-3GD will cost $415, the GV-N670OC-4GD will cost $550, and the GV-N680OC-4GD will cost (an astounding) $800.
I’m hoping that the rumored prices are in error and will be adjusted once the cards are available. These are neat cards that look to have plenty of cooling, but I’m still trying to figure out just what these cards have to offer to justify the huge jump over reference pricing. And, no, the superfluous gold plated HDMI connectors do not count. [For example, the 4GB Galaxy GTX 670 we recently reviewed was only $70 over reference while the Gigabyte card is rumored to be $150!]
The Gigabyte N66TOC-3GD factory overclocked GPU.
You can find links to the Gigabyte product pages in the charts above. If you have not already, please check out our GTX 660 Ti, GTX 670, and GTX 680 graphics card reviews for the full scoop on the various Kepler iterations. And if you are considering the Gigabyte N680OC-4GD, you should probably check out the dual GPU GTX 690 review as well (heh).
Subject: Systems | October 2, 2012 - 07:38 PM | Tim Verry
Tagged: maingear, kepler, Ivy Bridge, gtx 680m, gaming laptop
Maingear is a company that seemingly ascribes to the “go big or go home” motto, and nowhere is that sentiment made clearer than its latest gaming notebook: the Nomad 17.
Perhaps, the term “notebook” is a bit of an understatement here. The Nomad 17 is a 16.85” x 11.34” x 2.17” gaming notebook that packs the latest and greatest mobile technology into a package that is sure to give your back a workout should you attempt to use this beast as your daily driver (as someone that has attempted such a feat, I can attest to that heh). The Nomad 17 starts at $1,599 and goes up from there, but you do get a lot of hardware for the money.
An Intel Ivy Bridge Core i7-3840QM is the highest end CPU you can add, and it is also loaded out with either a NVIDIA GTX 675M or the GTX 680M graphics card and Optimus graphics switching technology. In addition, the Nomad can be configured with either two 512GB SSDs or two 750GB mechanical hard drives in a RAID O or RAID 1 array. The gaming laptop also does not skimp on RAM, allowing up to 32GB of DDR3 running at 1600MHz.
On the outside, you are getting a backlit keyboard, multitouch touchpad, and large 17” LED backlit display with matte anti-glare coating and a resolution of 1920x1080. On the audio front, it supports the THX TruStudio Pro audio codec and sports two speakers and a subwoofer by DynAudio. Connectivity options include a SD card reader, 6x Blu-ray burner/8x DVD writer optical drive, 802.11b/g/n Wi-Fi, and Bluetooth. External IO ports include one HDMI, one DVI, three USB 3.0, two USB 2.0, one Firewire, one optical audio out (S/PDIF), one Gigabit Ethernet/RJ45 port, and one RJ-11 port (of all things). Also, it features analog audio outputs, eSATA, and a VGA output.
The Nomad 17 with come pre-loaded with the 64-bit versions of either the Windows 7 Home, Premium, or Ultimate operating system.
But, the big reveal for gamers wanting to show off their gaming hardware is this: the Nomad 17 will be available in one of six custom, hand painted designs using glossy automotive paint.
The Nomad 17 is available now, and starts at $1,599. When decked out with the Core i7-3840QM, 4GB GTX 680M, 32GB system RAM, and two 512GB Crucial M4 SSDs (in RAID 0) mentioned above, the system total came out to $3,802. At that price, serious gamers only need apply, but is still an awesome piece of gaming technology nonetheless. Maingear has definitely packed the 17” laptop to the max with hardware.
You can find more photos of the Nomad 17 over at the Maingear website.
Subject: Graphics Cards | September 13, 2012 - 05:09 PM | Jeremy Hellstrom
Tagged: nvidia, msi, kepler, gtx 660, gk106, geforce, evga, factory overclocked
As those of you who have already read the post below this one know, ASUS decided to create a DirectCU II model for their GTX 660, with the famous heatpipe bearing heatsink. They have overclocked the GPU already and the card comes with tools to allow you to push it even further if you take the time to get to know your card and what it can manage. Check the full press release below.
Fremont, CA (September 13, 2012) - ASUS is excited to release the ASUS GeForce GTX 660 DirectCU II series featuring the Standard, OC and TOP editions. Utilizing the latest 28nm NVIDIA Kepler graphics architecture, the OC and TOP cards deliver a factory-overclock while all three cards feature ASUS exclusive DirectCU thermal design and GPU Tweak tuning software to deliver a quieter, cooler, faster, and more immersive gameplay experience. The ASUS GeForce GTX 660 DirectCU II series set a new benchmark for exceptional performance and power efficiency in a highly affordable graphics card. The ASUS GeForce GTX 660 DirectCU II is perfect for gamers looking to upgrade from last-generation graphics technology while retaining ASUS’ class-leading cooling and acoustic performance.
Superior Design and Software for the Best Gaming Experience ASUS equips the GeForce GTX 660 DirectCU II series with 2GB of GDDR5 memory clocked up to 6108MHz. The TOP edition features a blistering GPU core boost clock of 1137MHz, 104MHz faster than reference designs while the OC edition arrives with a factory-set GPU core boost speed of 1085MHz. Exclusive ASUS DIGI+ VRM digital power delivery and user-friendly GPU Tweak tuning software allows all cards to easily overclock beyond factory-set speeds offering enhanced performance in your favorite game or compute intensive application.
The ASUS GeForce GTX 660 DirectCU II series feature exclusive DirectCU technology. The custom designed cooler uses direct contact copper heatpipes for faster heat transduction and up to 20% lower normal operating temperatures than reference designs. The optimized fans are able operate at lower speeds providing a much quieter gaming or computing environment. For enhanced stability, energy efficiency, and overclocking margins the cards feature DIGI+ VRM digital power deliver plus a class-leading six-phase Super Alloy Power design for the capacitors, chokes, and MOSFETs meant to extend product lifespan and durability while operating noise-free even under heavy workloads.
ASUS once again includes the award winning GPU Tweak tuning suite in the box. Overclocking-inclined enthusiasts or gamers can boost clock speeds, set power targets, and configure fan operating parameters and policies; all this and more is accessible in the user-friendly interface. GPU Tweak offers built-in safe guards to ensure all modifications are safe, maintaining optimal stability and card reliability.
Subject: Graphics Cards | September 13, 2012 - 04:49 PM | Jeremy Hellstrom
Tagged: nvidia, msi, kepler, gtx 660, gk106, geforce, evga
The non-Ti version of the GTX 660 has arrived on test benches and retailers, with even the heavily overclocked cards being available at $230, like EVGA's Superclocked model or MSI's OC'd card once you count the MIR. That price places it right in between the HD 7850 and 7870, and ~$70 less than the GTX 660 Ti, while the performance is mostly comparable to a stock HD7870 though the OC versions can top the GTX660.
[H]ard|OCP received ASUS' version of the card, a DirectCU II based version with the distinctive heatpipes. ASUS overclocked the card to a 1072MHz base clock and 1137MHz GPU Boost and [H] plans to see just how much further the frequencies can be pushed at a later date. Their final word on this card for those looking to upgrade, for those of you with "a GTX 560 Ti, and even the GTX 570, the GTX 660 is an upgrade".
"NVIDIA is launching the new GeForce GTX 660 GPU, codenamed GK106. We have a retail ASUS GeForce GTX 660 DirectCU II custom video card fully evaluated against a plethora of competition at this price point. This brand new GPU aims for a price point just under the GTX 660 Ti but still promises to deliver exceptional 1080p gaming with AA."
Here are some more Graphics Card articles from around the web:
- Nvidia's GeForce GTX 660 @ The Tech Report
- ASUS GTX 660 Direct CU II TOP Review @ OCC
- NVIDIA GeForce GTX 660 Launch Review @ Neoseeker
- EVGA GeForce GTX 660 SC (SuperClocked) 2GB @ Bjorn3D
- Nvidia GeForce GTX 660 @ Hardware.info
- NVIDIA Geforce GTX 660 Reviews @Hi Tech Legion
- The NVIDIA GeForce GTX 660 Review: GK106 Fills Out The Kepler Family @ AnandTech
- SI GEFORCE GTX 660 Twin Frozr 2GB OC @ Tweaktown
- Gigabyte GeForce GTX 660 @ Legion Hardware
- Gigabyte GTX 660 Overclock 2GB Graphics Card Review @ eTeknix
- EVGA GeForce GTX 660 2GB SuperClocked @ Benchmark Reviews
- MSI GTX 660 OC Edition Twin Frozr @ Kitguru
- Nvidia GeForce GTX 660 @ Techspot
- Gigabyte GTX 660 OC Video Card Review @ Ninjalane
- MSI GTX 660 Twin Frozr 2GB OC @ LanOC Reviews
- NVIDIA GeForce GTX 660 Overclocked Graphics Card Review (EVGA/ZOTAC)@ HardwareHeaven
- EVGA GTX 660 Superclocked 2Gb @ LanOC Reviews
- NVIDIA GeForce GTX 660 Review @ Hardware Canucks
- ASUS, KFA2 and MSI GeForce GTX 660 reviews with 2-way SLI @ Guru of 3D
- MSI GeForce GTX 660 Twin Frozr 2 GB @ techPowerUp
- ZOTAC GeForce GTX 660 2 GB @ techPowerUp
- Gigabyte GTX 660 Windforce OC 2 GB @ techPowerUp
- ASUS GeForce GTX 660 Direct Cu II 2 GB @ techPowerUp
NVIDIA GeForce GTX 660 Video Card Review w/ MSI and EVGA @ Legit Reviews
- Six GeForce GTX 660 Ti graphics cards: ASUS, EVGA, Gigabyte, MSI and Zotac @ Hardware.info
- Gigabyte GTX 660 Ti OC Windforce @ Kitguru
- AFOX Radeon HD 7850 (Single Slot), MSI R7870 Hawk Graphics Cards @ iXBT Labs
- Inno3D GTX 680 iChill Black Series Accelero Hybrid 4GB Overclocked @ Tweaktown
- MSI Geforce GTX 670 Power Edition @ Rbmods
- i3DSpeed, August 2012 @ iXBT Labs
- Arctic Accelero Xtreme 7970 VGA Cooler Review @ eTeknix
- Sapphire Radeon HD 7970 Vapor-X OC 6GB Graphics Card Review @ eTeknix
- Sapphire FleX HD 7770 GHz Edition @ LanOC Reviews
GK106 Completes the Circle
The release of the various Kepler-based graphics cards have been interesting to watch from the outside. Though NVIDIA certainly spiced things up with the release of the GeForce GTX 680 2GB card back in March, and then with the dual-GPU GTX 690 4GB graphics card, for quite quite some time NVIDIA was content to leave the sub-$400 markets to AMD's Radeon HD 7000 cards. And of course NVIDIA's own GTX 500-series.
But gamers and enthusiasts are fickle beings - knowing that the GTX 660 was always JUST around the corner, many of you were simply not willing to buy into the GTX 560s floating around Newegg and other online retailers. AMD benefited greatly from this lack of competition and only recently has NVIDIA started to bring their latest generation of cards to the price points MOST gamers are truly interested in.
Today we are going to take a look at the brand new GeForce GTX 660, a graphics cards with 2GB of frame buffer that will have a starting MSRP of $229. Coming in $80 under the GTX 660 Ti card released just last month, does the more vanilla GTX 660 have what it takes to replace the success of the GTX 460?
The GK106 GPU and GeForce GTX 660 2GB
NVIDIA's GK104 GPU is used in the GeForce GTX 690, GTX 680, GTX 670 and even the GTX 660 Ti. We saw the much smaller GK107 GPU with the GT 640 card, a release I was not impressed with at all. With the GTX 660 Ti starting at $299 and the GT 640 at $120, there was a WIDE gap in NVIDIA's 600-series lineup that the GTX 660 addresses with an entirely new GPU, the GK106.
First, let's take a quick look at the reference card from NVIDIA for the GeForce GTX 660 2GB - it doesn't differ much from the reference cards for the GTX 660 Ti and even the GTX 670.
The GeForce GTX 660 uses the same half-length PCB that we saw for the first time with the GTX 670 and this will allow retail partners a lot of flexibility with their card designs.
Subject: Graphics Cards | September 13, 2012 - 09:38 AM | Tim Verry
Tagged: nvidia, kepler, gtx 650, graphics cards, geforce
Ah, Kepler: the (originally intended as) midrange graphics card architecture that took the world by storm and allowed NVIDIA to take it from the dual-GPU GeForce GTX 690 all the way down to budget discrete HTPC cards. So far this year we have seen the company push Kepler to its limits by adding GPU boost and placing it in the GTX 690 and GTX 680. Those cards were great, but commanded a price premium that most gamers could not afford. Enter the GTX 670 and GTX 660 Ti earlier this year and Kepler started to become an attractive option for gamers wanting a high-end single GPU system without breaking the bank. Those cards, at $399 and $299 respectively were a step in the right direction to making the Kepler architecture available to everyone but were still a bit pricey if you were on a tighter budget for your gaming rig (or needed to factor in the Significant Other Approval Process™).
Well, Kepler has now been on the market for about six months, and I’m excited to (finally) announce that NVIDIA is launching its first Kepler-based budget gaming card! The NVIDIA GeForce GTX 650 brings Kepler down to the ever-attractive $109 price point and is even capable of playing new games at 1080p above 30FPS. Not bad for such a cheap card!
With the GTX 650, you are making some sacrifices as far as hardware, but things are not all bad. The card features a mere 384 CUDA cores and 1GB of GDDR5 memory on a 128-bit bus. This is a huge decrease in hardware compared to the GTX 660 Ti’s 1344 CUDA cores and 2GB memory on a 192-bit bus – but that card is also $200 more. And while the GTX 650 runs the memory at 5Gbps, NVIDIA was not shy about pumping up the GPU core clockspeed. No boost functionality was mentioned but the base clockspeed is a respectable 1058 MHz. Even better, the card only requires a single 6-pin PCI-E power connector and has a TDP of 64W (less than half of its higher-end GeForce brethren).
The following chart compares the specifications between the new Geforce GTX 650 through the GTX 670 graphics card.
Click on the above chart for a larger image.
The really important question is how well it handles games, and NVIDIA showed off several slides with claimed performance numbers. Taking these numbers with a grain of salt as they are coming from the same company that built the hardware, the GTX 650 looks like a capable GPU for the price. The company compared it to both its GTS 450 (Fermi) and AMD’s 7750 graphics card. Naturally, it was shown in a good light in both comparisons, but nothing egregious.
NVIDIA is claiming an 8X performance increase versus the old 9500 GT, and an approximate 20% speed increase versus the GTS 450. And improvements to the hardware itself has allowed NVIDIA to improve performance while requiring less power; the company claims the GTX 650 uses up to half the power of its Fermi predecessor.
The comparison between the GTX 650 and AMD Radeon HD 7750 is harder to gauge, though the 7750 is priced competitively around the GTX 650’s $109 MSRP so it will be interesting to see how that shakes out. NVIDIA is claiming anywhere from 1.08 to 1.34 times the performance of the 7750 in a number of games, shown in the chart below.
If you have been eyeing a 7750, the GTX 650 looks like it might be the better option, assuming reviewers are able to replicate NVIDIA’s results.
Keep in mind, these are NVIDIA's numbers and not from our reviews.
Unfortunately, NVIDIA did not benchmark the GTS 450 against the GTX 650 in the games. Rather, they compared it to the 9500 GT to show the upgrade potential for anyone still holding onto the older hardware (pushing the fact that you can run DirectX 11 at 1080p if you upgrade). Still, the results for the 650 are interesting by themselves. In MechWarrior Online, World of Warcraft, and Max Payne 3 the budget GPU managed at least 40 FPS at 1920x1080 resolution in DirectX 11 mode. Nothing groundbreaking, for sure, but fairly respectable for the price. Assuming it can pull at least a min of 30 FPS in other recent games, this will be a good option for DIY builders that want to get started with PC gaming on a budget.
All in all, the NVIDIA GeForce GTX 650 looks to be a decent card and finally rounds out the Kepler architecture. At this price point, NVIDIA can finally give every gamer a Kepler option instead of continuing to rely on older cards to answer AMD at the lower price points. I’m interested to see how AMD answers this, and specifically if gamers will see more price cuts on the AMD side.
If you have not already, I strongly recommend you give our previous Kepler GPU reviews a read through for a look at what NVIDIA’s latest architecture is all about.
PC Perspective Kepler-based GTX Graphics Card Reviews:
- GeForce GTX 690: Dual GK104 Kepler Greatness
- GeForce GTX 680: Kepler is ready for retail
- GeForce GTX 670: Kepler for $399
- GeForce GTX 660 Ti: Another GK104 Option for $299
- GeForce GTX 660: GK106 Completes the Circle
Multiple Contenders - EVGA SC
One of the most anticipated graphics card releases of the year occurred this month in the form of the GeForce GTX 660 Ti from NVIDIA, and as you would expect we were there on the day one with an in-depth review of the card at reference speeds.
The GeForce GTX 660 Ti is based on GK104, and what you might find interesting is that it is nearly identical to the specifications of the GTX 670. Both utilize 7 SMX units for a total of 1344 stream processors – or CUDA cores – and both run at a reference clock speed of 915 MHz base and 980 MHz Boost. Both include 112 texture units though the GeForce GTX 660 Ti does see a drop in ROP count from 32 to 24. Also, L2 cache drops from 512KB to 384KB along with a memory bus width drop from 256-bit to 192-bit.
We already spent quite a lot of time talking about the GTX 660 Ti compared to the other NVIDIA and AMD GPUs in the market in our review (linked above) as well as on our most recent episode of the PC Perspective Podcast. Today's story is all about the retail cards we received from various vendors including EVGA, Galaxy, MSI and Zotac. We are going to show you each card's design, the higher clocked settings that were implemented, performance differences between them and finally the overclocking comparisons of all four.