Zbox mini PCs get Ivy Bridge Core i3, i5 upgrades

Zotac has beefed up its Zbox lineup with two new models based on Ivy Bridge processors. Until now, the company’s most powerful mini PC has been based on the Core i3-2330M, a 2.2GHz duallie from the Sandy Bridge generation. The new Zbox ID88 boasts a dual-core Core i3-3220 clocked at 2.8GHz, while the ID89 sports a Core i5-3470 with a 2.9GHz base frequency and a 3.6GHz Turbo peak. Both chips can execute four threads in parallel via Hyper-Threading. However, only the i5 includes AES acceleration capabilities and some extra virtualization mojo.

Apart from their processors, the new Zboxes appear to be identical. Intel’s H61 Express chipset anchors the underlying motherboards, which have dual SO-DIMM slots and support for up to 16GB of RAM. Although the H61 platform doesn’t have integrated USB 3.0, Zotac appears to have added an auxiliary controller to provide the systems with dual SuperSpeed ports. There are dual USB 2.0 ports, as well, plus two Gigabit Ethernet jacks, a memory card reader, and plenty of audio and video outs. Wireless options include 802.11n Wi-Fi and Bluetooth 4.0.

The Zbox chassis has a single 2.5" hard drive bay, which is empty in barebones models but comes loaded with a 500GB mechanical drive in the Plus variants of each system. Those configs also include 4GB of RAM, but you’ll still have to provide the operating system yourself.

As usual, Zotac includes a bundle of goodies with the new Zbox systems. The best part is arguably the MCE-compatible IR remote, which plays nicely with XBMC. Zotac also throws in a stand for propping up the chassis vertically and a VESA mount for strapping the chassis onto the back of a monitor. Only some monitors will support the VESA mount, but the bracket can be placed anywhere you can sink a few screws.

While the new Zboxes have much faster CPUs than their predecessors, their graphics horsepower is a little lacking. Unlike the ID42 we reviewed in February, these machines don’t feature discrete GeForce graphics. Instead, you’re stuck with the Intel HD Graphics 2500 IGP, which should be fine for everything but serious gaming. The ID42 is saddled with a pokey 1.1GHz Celeron CPU, so it entails some compromise, too.

We don’t have US pricing for the new Zboxes yet, but Zotac tells us the ID88 and ID89 will sell in Germany for €329 and €399, respectively, with the Plus variants each carrying a €80 premium. A direct exchange-rate conversion pegs US pricing at $432 for the ID88 and $524 for the ID89. I suspect actual prices will be a tad lower stateside, if only to slip under the $400 and $500 thresholds.

Update: Zotac has provided us with US pricing for the new machines, which will indeed be cheaper on this side of the Atlantic—quite a lot cheaper, in fact. The ID88 is set to retail for only $340, and the ID89 will be priced at $430. You can expect to add $110 for the Plus version of the ID88 and $100 for the equivalent ID89.

Comments closed
    • Hinton
    • 6 years ago

    [i<]"Zotac has provided us with US pricing for the new machines, which will indeed be cheaper on this side of the Atlantic—quite a lot cheaper, in fact."[/i<] Why would they be cheaper? As far as I can tell they're priced more or less identical (hint. In most countries, including Germany, it's not legal to quote any other price than what the consumer is to pay. That obviously includes VAT.)

    • gmskking
    • 6 years ago

    Nice form factor, but too expensive still.

    • rwburnham
    • 6 years ago

    I recently bought a Zbox with an AMD E2-1800 APU, and it handles local media just fine, even full HD content. Problem is, it cannot handle HD Netflix streaming. That’s fine for me, as I have consoles for that, but some people might be put off by that fact.

      • TO11MTM
      • 6 years ago

      I blame AMD’s drivers… I’ve seen these handle Netflix streaming on HD without any issues…. High CPU though (60-80%)

      Maybe I just missed it dropping frames?

        • Deanjo
        • 6 years ago

        Don’t blame AMD’s drivers, blame silverlight that does not support AMD video rendering acceleration.

    • HisDivineOrder
    • 6 years ago

    Too bad Zotac is always a year behind the current CPU here. Other than them getting these IB CPU’s for a steal, I think the user would have been better served by them waiting a month and using a Haswell variant instead for a system this size and so reliant on the IGP.

    • mcnabney
    • 6 years ago

    When i3 laptops are going under $400 with Win8 I am having a hard time in seeing the value in this device. So as a Linux box it is $450 with a drive. Not impressed.

    • Chrispy_
    • 6 years ago

    The Intel IGP limitation reduces the appeal of these things.

    If Zotac offered a A10-4600M or A8-4500M (both 35W processors) they’d be much better at gaming, whilst still being perfectly adequate at 1080p decoding and web-browsing.

    Edit – I should add that the 35W TDP is the same as the Core i3-3220T which this thing has in it.

      • chuckula
      • 6 years ago

      For once I will say this without trolling: AMD really would be better if the only Intel chips they offer are stuck with HD 2500. Now, if they could have used HD 4000 parts with 35 Watt TDPs, I would call it a draw because the mobile AMD parts really aren’t substantially better at those power envelopes. Haswell’s improved GPU will have the biggest impact in systems like these.

        • Chrispy_
        • 6 years ago

        Indeed, Haswell will briefly oust Mobile Trinity from the IGP throne, but running things on my HD2000 IGP compared to my GTX650M shows endless rendering discrepancies that prove Intel’s IGP performance figures are still not a valid comparison, because they’re not doing all of the work; That’s “cheating” in my book.

          • rootheday3
          • 6 years ago

          If you don’t mind my asking, what specific “endless rendering issues” are you seeing and what Intel drivers are you running on.

          Disclosure: I work in the Intel graphics driver team and I can assure you that we have no intention of “cheating” and are working aggressively to fix issues. If you can give me a concrete bug report on recent/current driver, I will get it into our bug database.

            • chuckula
            • 6 years ago

            [quote<] Disclosure: I work in the Intel graphics driver team and I can assure you that we have no intention of "cheating" and are working aggressively to fix issues. If you can give me a concrete bug report on recent/current driver, I will get it into our bug database.[/quote<] You do realize that it is my greatest hope to post a troll that is so over the top that you actually print it out and post it by the water cooler right? P.S. --> I'm a big Linux guy and I appreciate the open-source drivers and the early support of Haswell on Linux. Pass that on to anyone you may know on that side of the fence.

            • rootheday3
            • 6 years ago

            Your trolling is an excellent blend of finely diced fanboi rage, served over a bed of ripe “forum wisdom”, sweetened with a delicious yet subtle flavor of sarcasm and irony. Many who rush through the meal seem to miss that last part and thus fail to appreciate the trolling as it is intended. I lift my hat to the great writing- keep it up and I will tell you when I print one.

            • Chrispy_
            • 6 years ago

            Well, the only laptops I have with Intel IGP’s in them are Sandy Bridge. They have so many draw issues with CAD packages like Microstation, GC, Rhino and even some of the basic DWG viewers that we just can’t work with them. Whether the fault of this is Autodesk/Bentley/McNeel or Intel I don’t know, but I do know that Geforce/Quadro/Radeon/FirePro work without problems.

            Recent/current drivers are unlikely to be installed on those laptops but my gaming laptop is on 15.28 and I know for sure that the driver defaults button drops the texture filtering and aniso settings right down to wholly unacceptable levels when AMD and Nvidia’s defaults are at full quality.

            Already, your latest drivers have abandoned quality/performance improvements for IGP’s that are barely two years old – the 15.31 drivers only cover HD4000/2500, whilst both AMD and Nvidia are making driver changes that improve entire generations of cards – not just the current generation.

            So this is what millions of people will be stuck with forevermore:
            [url<]https://techreport.com/review/21099/amd-a8-3500m-fusion-apu/11[/url<] [url<]http://www.anandtech.com/Gallery/Album/1890#13[/url<] Thanks. Having seen HD4000 in the flesh it's a big improvement but the compatibility issues last year at launch were so great that [url=http://semiaccurate.com/2012/04/23/comparing-the-ends-of-the-spectrum/<]you were only listing games that [i<]did[/i<] work[/url<], rather than listing those few games with known issues. Assuming that Intel is actually serious about being a third competitor in 3D graphics, the results will come out in the reviews when Haswell launches. Things we take for granted with competing products will be seen in a negative light if Intel fails to provide them - be it driver defaults, non-gaming (OpenGL or OpenCL) compatibility, IQ settings in the driver control panel, or maximum frames/second at the expense of uneven frame delivery etc. Despite that largely critical post, I have to commend the Intel driver team for making such massive progress. Even as recently as three years ago, Intel IGPs were so farcical that something as simple as launching many games was off-limits. Here we are today with a mainstream Intel graphics product that can actually deliver acceptable performance in a great many cases.

            • rootheday3
            • 6 years ago

            thanks for the extensive post.

            For workstation apps, I will have to check with the folks who support those to get latest/current status.

            Re the support windows – it is true that Intel has ~2 year window of active supports on the mainline of our driver development. I would of course love for it to be more but we’ve been stretched a lot to aggressively ramp up on feature sets (APIs, display technologies), app compatibility, operating systems (see Android announcement at CES and Linux transcoding at NAB recently), etc while expanding both up to higher performance tiers and lower power/higher efficiency. There is only so much polishing possible on the older products vs optimizing the upcoming products-> the anisotropic filtering quality difference between HD 2000/3000 vs Llano is an example – that was a hardware limitation of the texture sampler. No amount of driver updates can make that better. We did fix the sampler hardware in IvyBridge (HD2500/HD4000) to have “perfect” circles in the aniso test.

            A key piece of this is that we have been basically doing a major graphics architecture update in each generation for the last 3 years. NVidia and AMD typically have generations last much longer and have much smaller differences between generations (AMD VLIW5->VLIW4->GCN is the counter example that proves the rule – took them a while to get that right/tuned). As a consequence it is easier for them to have a longer support tail. I would ask however – how many recent driver updates from AMD or NVidia do they advertise as offering performance or functional fixes for products that are older than a year or so? They may still be “in the driver inf” as supported, but it doesn’t mean those older driver code paths are being actively improved or maintained.

            [quote<]I know for sure that the driver defaults button drops the texture filtering and aniso settings right down to wholly unacceptable levels when AMD and Nvidia's defaults are at full quality.[/quote<] Not sure what you mean by this. The driver defaults should be set to "application specified" - meaning no driver override; driver uses what the app asks for. If you can give me the driver build #, I can see if it got screwed up somehow. The control panel overrides on Aniso are to force aniso ON for apps that only offer bilinear/trilinear-> should be something that you selectively turn on for apps that don't offer in-game option for aniso. That said, the HD2000 texture sampler aniso quality is what it is.... Re Texture quality -> this option never made sense - it just did LOD biasing +1/-1. I argued successfully to have it removed and It should be gone from 15.28 drivers and later. I understand that we will be judged on results and quality. I only ask that people don't come prejudging based on older products and buggy/limited drivers from years ago. The dreaded "list of games page"... so sadly misunderstood. It was never supposed to be a "these are the only games that are compatible" - it was "these are ones that we have tested in our lab AND believe will offer a playable gaming experience" with the additional caveat that marketing only published one page covering HD2500 AND HD4000... We actually tested many more titles AND many more were playable on HD4000... sigh. That said, I will tell you that we find a LOT of bugs in games - in some cases we can work around them in the driver but in many cases we can't. We pass these back to the ISVs - in many cases however they are either too busy or not motivated to issue a patch with fix for older games. We could use your help as advocates to the ISVs - tell them on their forums that you want them to work with us to ensure that apps work well on the Intel graphics devices you already own or intend to buy.. To prevent future games from having these sorts of issues, we have a large team of AEs working with game developers to test/tune games against our current and future hardware; in the last couple years we have tested ~250 to 300 titles pre-release on multiple skus at various settings, in many cases through multiple rounds (alpha/beta/gold)

            • Chrispy_
            • 6 years ago

            I’ll get back to you on the driver defaults, I remember the defaults button putting all the sliders back down to performance and the texture quality dropping down to superblurred again. I won’t be home until the weekend to check but in terms of build number I believe it’s the one that’s been on the downloads page since mid-March.

            It does genuinely sound like Intel are making a serious effort to provide compelling IGP’s.
            Nobody is really going to be able to game on a Sandy IGP, and Ivy’s HD2500 lacks the horsepower to run current titles, but my few experiences with HD4000 left me satisfied that they were finally good enough for casual gamers. As long as you guys maintain this momentum I think Intel graphics will go far.

            Let’s just hope you guys don’t kill AMD in the process – we’re already seeing the effects of no high-end competition from a CPU perspective 😉

            • sschaem
            • 6 years ago

            Where were you 10 years ago ! 🙂

            I’m glad you guys are pro-active and roaming the web. But in the badlands, your professionalism will be test ….

    • UberGerbil
    • 6 years ago

    Looking forward to the Haswell generation of these things.

      • Deanjo
      • 6 years ago

      It would be interesting to see but I would only be interested in Haswell if they finally fixed intels video frame rate issue with accelerated decoding. There hasn’t been much published about its video decode abilities.

    • Deanjo
    • 6 years ago

    [quote<]The best part is arguably the MCE-compatible IR remote, which plays nicely with XBMC.[/quote<] Complete and utter crap remote on anything outside of windows (pretty crappy in windows as well). If you are going to run the superior linux XBMC you are going to want to replace it with a good MCE remote or even better get the Pulse 8 HDMI-CEC USB adapter and use your regular TV or amp remote. [url<]http://www.pulse-eight.com/store/products/104-usb-hdmi-cec-adapter.aspx[/url<]

      • rwburnham
      • 6 years ago

      If it’s the same remote that comes with the AMD E2-1800 APU model, it’s crap. I tried using the remote and it would move two spaces when I pressed the direction button once in WMC.

        • Deanjo
        • 6 years ago

        They use the same remotes in all their products.

    • OneArmedScissor
    • 6 years ago

    [quote<]While it's also a dual-core chip, the i5 can execute four threads in parallel via Hyper-Threading.[/quote<] ??? So do all i3s, mobile or desktop. They just don't get turbo, which means gimped clock speeds in today's world.

      • Deanjo
      • 6 years ago

      No i3’s have turbo but they do have Hyperthreading.

        • Vulk
        • 6 years ago

        [url<]http://ark.intel.com/products/71053/[/url<] HT, no turbo.

          • Deanjo
          • 6 years ago

          That’s what I said.

          There is a difference between:

          No i3’s have turbo….

          and

          No, i3’s have turbo….

      • Dissonance
      • 6 years ago

      Fixed.

    • dpaus
    • 6 years ago

    These would also make first-rate firewall appliances when used with something like [url=http://www.untangle.com<]Untangle (TM)[/url<]

      • Deanjo
      • 6 years ago

      Wonder if they are intel NICs or more realtek crap. (BTW, pfsense all the way!)

      • GeneS
      • 6 years ago

      Too subtle, try again.

      • NeelyCam
      • 6 years ago

      Good job, although I let you have this one

        • dpaus
        • 6 years ago

        They run out of coffee at the death metal bar?

          • NeelyCam
          • 6 years ago

          They don’t serve coffee, and they ran out of espresso stout

      • strikeleader
      • 6 years ago

      I would prefer [url=http://www.sophos.com/en-us/products/free-tools/sophos-utm-home-edition.aspx/<]Sophos UTM9[/url<] for the firewall.

Pin It on Pinterest

Share This