IT’s a dual PSU, platinum. PWS-920P-SQ, but one is disconnected.
I’m keeping an eye on titanium, but they usualy are rated at massive power. On the other hand, the spec incudes efficiency at 10% load, so it might work 
1 Like
PWS-920P-SQ
Ahh cool. I’m not fluent in Supermirco modelnumbers, and on the top of my head, V3/V4 was the tippingpoint for some manufacturers to switch from gold rated PSUs from gold to platinum.
My old job had Cisco boxes with both efficiencies.
2% great efficiency at 20% load on baseload of 180 watts is still 5ish watts - very inline with the numbers I saved (however I did go from gold → titanium). Weather or not this is worthwhile the capital investment on a PSU, I don’t know. My 2u box was upgraded to a Seasonic Prime PX-450 - it’s a seriously nice unit
I think Supermicro made the switch earlier, or perhaps, staggered depending on the product – more power hungry products got better power supplies.
My contraption is originally SuperStorage 6028R-E1CR12L with dual-socket v3, and it had a pair of platinum power supplies from the factory
It still baffles me what kind of value can you get for pennies: I bought it from a local electronic recycler few years ago, who harvests old crap from decommissioned data centers, and sells both parts and can build-to-order servers; this server had 2x E5-2680 V3 2.5Ghz 12 Core (Total 24 cores, 48 hyperthreaded), 128 GB of ram, SAS3 backplane with support for NVME on 4 ports, and Supermicro’s 3008-based HBA… From the looks of it no human touched the interior after it was built by supermicro.
For how much? $300…
After this, how can one justify buying consumer chassis, or motherboards, or processors, or anything for that matter?
Even if electricity is expensive, and power supplies cheap – it may or may not be: at 20%, 50%, and 100% load the efficiency difference with Platinum is just 1-2%
Titanium, however, qualitatively differs from all previous tiers – there is another anchor at 10% load. Since we are talking about low power home systems, and the server power supplies are pushing kilowatt capacity to start with – having this 10% anchor is very helpful. You can assume that the supply hits the targets only around those anchors – because that’s what standard requires, so if your server consumes 100W, and you with 1000W Titanium power supply you get promised efficiency, but with Platinum and lower – who knows: they are not rated there, and can perform worse. Ideally of course you want to be at 50% load – that’s where any PSU performs the best.
Well my consumer (fractal) case sits in a corner discretely and quietly and doesn’t anger my wife.
and the consumer cpu has an igpu that’s useful for transcoding video.
1 Like
Hehe Mine site outside of the living space, and also does not frustrate anyone. Computers should not be in the living quirters, if at all possible.
Why do you need to transcode video? Plex? Why not direct-play? QuckSync is fast, but does not yield a very good quality.
I like to store my disc rips as remuxes, but the Apple TV has issues with mkv’s and also often tone mapping, which leads to a transcode, and viewing remotely tends to need to transcode.
also the igpu can be used to accelerate immich’s AI photo learning features. I see it’s enabled although I don’t know if it’s any faster than cpu in practice.
I see. Mkv shall result in “Direct Stream” if the stream inside is compatible, and remuxing does not benefit from GPU either.
AppleTV can do tone mapping though, no? At least, mine does. (I have a very similar setup – plex, appletv, SDR projector)
Plex can do this: Creating Optimized Versions | Plex Support, which will yield much better quality – because you can do it offline, using different equipment, with better settings – and you do want to squeeze every ounce of quality when you are already bandwidth constrained. QuickSync would be doing you a disservice.
That way you get best of both worlds – “archiving” with obnoxious mkv capabilities, and playable compatible file for actual real world use, and without sacrificing any quality
That I know nothing about… but if it works – it works. I would use a real accelerator for that though, not something built in into the cpu by marketing people and given out for free
Side remark, on the topic why do I still think everyone shall stick to TrueNAS Core: it’s becase Scale is still yet to reach feature parity with a platform they deem obsolete and poorly supported, or whatever their marketing department decided. We are talking basic power managemetn. Ugh.
Replacing E5-2637 v4 with E5-1630 v4 (loses smart cache and QPI links but gains higher clock) helped shave off about 3W. Coupled with enabling ASPM,
I’m now at 160W idle:
So far: 212W → 160W, about $20/month savings. Cost: likely negative, once I sell the pile of hardware and processors I bought for testing of all this..
Next: replacing AOC-S3008L-L8e (high base power consumption, no ASPM) with 9500 series HBA (⅓ rd power consumption source, plus supports ASPM). I’ll report once I receive it.
1 Like
gonna be hard calculating the payoff window for that LSI 9500!
1 Like
Hehe I already did:
I bought it on ebay for $110 (New Broadcom HBA 9500-8i Tri-Mode Storage Cntroller SATA 6Gb/s SAS 12Gb/s PCIe | eBay $98+tax. The current one can be sold for about $20, but I’m not going to – I’ll keep it for another install, it’s an actually genuine supermicro one, unlike whatever the hell the Broadcom one was sourced from)
Assuming I’ll save about 10 watts, at 52 cents per kilowatt, 110/(10*24*30/1000*0.52)/12=2.44 years.
I understand that while it will definitely pay off more than once during the life of the server, it’s not much, but it’s now turned into sports for me – how much can I squeeze on the cheap without degrading peak performance? (because otherwise Atom C3000 based nas is an answer of course)
(Perhaps the next-next step would be replacing the expander backplane with a dumb one - I don’t know how much expander consumers, but it has a large heatsink and that is quite warm to the touch. I will lose NVME support – but I don’t use NVME in disk bays, so technically it’s not a loss of functionality)
I like following this project.
You could probably even go backplaneless, and use SATA/SAS SFF-8482 male to SATA female adapters. That would require you to have you PSU produce 12x SATApower connectors, but that should be easily manageable
I use SATA power splitters to get extra ports on my consumer case. Extra handy because I usually need to cut the 3.3v pin for SAS drives to spin up, and it’s much nicer cutting a cheap extension cord.
they also make breakout cables with molex power if you have molex plugs you want to put to use.
This gets messy very fast – in fact, backplane was a requirement from the beginning. I had a server before without – and replacing a disk was an ordeal. Open the case, disconnect wires, and that if you manage to identify the correct disk in the first place, unscrew the disk, and then what to do with SAS disks? Another bunch of adapters…
The expanders consume some power, but TQ backplanes without one only have very low power circuitry, extra few milliwatts are totally worth it for the massive convenience in the form of disk swaps and disk identification.
There are toolless backplanes – like the one 45Drives designed for their servers, this would be another alternative.
Totally. Those molex splitters are abundant and cheap.
On the related news:
HBA arrived from one place, cable arrived from the other. Put it all together, and:
some horseshit ensued
/var/log/messages.0.bz2-Jul 10 23:39:54 truenas.arrogantrabbit.com mpr0: Controller reported scsi ioc terminated tgt 25 SMID 598 loginfo 31120303
/var/log/messages.0.bz2:Jul 10 23:39:54 truenas.arrogantrabbit.com (da3:mpr0:0:19:0): CAM status: CCB request completed with an error
/var/log/messages.0.bz2-Jul 10 23:39:54 truenas.arrogantrabbit.com (da6:mpr0:0:22:0): Retrying command, 3 more tries remain
/var/log/messages.0.bz2-Jul 10 23:39:54 truenas.arrogantrabbit.com (da6:mpr0:0:22:0): WRITE(10). CDB: 2a 00 02 00 02 a0 00 00 e0 00
/var/log/messages.0.bz2-Jul 10 23:39:54 truenas.arrogantrabbit.com (da3:mpr0:0:19:0): Retrying command, 3 more tries remain
/var/log/messages.0.bz2-Jul 10 23:39:54 truenas.arrogantrabbit.com mpr0: Controller reported scsi ioc terminated tgt 21 SMID 604 loginfo 31120303
/var/log/messages.0.bz2-Jul 10 23:39:54 truenas.arrogantrabbit.com mpr0: Controller reported scsi ioc terminated tgt 21 SMID 602 loginfo 31120303
/var/log/messages.0.bz2-Jul 10 23:39:54 truenas.arrogantrabbit.com mpr0: Controller reported scsi ioc terminated tgt 27 SMID 606 loginfo 31120303
/var/log/messages.0.bz2-Jul 10 23:39:54 truenas.arrogantrabbit.com mpr0: Controller reported scsi ioc terminated tgt 27 SMID 605 loginfo 31120303
/var/log/messages.0.bz2-Jul 10 23:39:54 truenas.arrogantrabbit.com mpr0: Controller reported scsi ioc terminated tgt 24 SMID 607 loginfo 31120303
/var/log/messages.0.bz2-Jul 10 23:39:54 truenas.arrogantrabbit.com mpr0: Controller reported scsi ioc terminated tgt 24 SMID 603 loginfo 31120303
/var/log/messages.0.bz2:Jul 10 23:39:54 truenas.arrogantrabbit.com (da6:mpr0:0:22:0): CAM status: CCB request completed with an error
/var/log/messages.0.bz2-Jul 10 23:39:54 truenas.arrogantrabbit.com (da3:mpr0:0:19:0): WRITE(10). CDB: 2a 00 02 00 02 a0 00 00 e0 00
/var/log/messages.0.bz2-Jul 10 23:39:54 truenas.arrogantrabbit.com mpr0: Controller reported scsi ioc terminated tgt 18 SMID 610 loginfo 31120303
/var/log/messages.0.bz2:Jul 10 23:39:54 truenas.arrogantrabbit.com (da3:mpr0:0:19:0): CAM status: CCB request completed with an error
/var/log/messages.0.bz2-Jul 10 23:39:54 truenas.arrogantrabbit.com mpr0: Controller reported scsi ioc terminated tgt 18 SMID 609 loginfo 31120303
/var/log/messages.0.bz2-Jul 10 23:39:54 truenas.arrogantrabbit.com (da3:mpr0:0:19:0): Retrying command, 3 more tries remain
/var/log/messages.0.bz2-Jul 10 23:39:54 truenas.arrogantrabbit.com (da9:mpr0:0:25:0): WRITE(16). CDB: 8a 00 00 00 00 09 18 5f fc a0 00 00 00 e0 00 00
/var/log/messages.0.bz2-Jul 10 23:39:54 truenas.arrogantrabbit.com (da6:mpr0:0:22:0): Retrying command, 3 more tries remain
/var/log/messages.0.bz2-Jul 10 23:39:54 truenas.arrogantrabbit.com (da0:mpr0:0:16:0): WRITE(16). CDB: 8a 00 00 00 00 09 18 5f fc a0 00 00 00 e0 00 00
/var/log/messages.0.bz2:Jul 10 23:39:54 truenas.arrogantrabbit.com (da9:mpr0:0:25:0): CAM status: CCB request completed with an error
/var/log/messages.0.bz2-Jul 10 23:39:54 truenas.arrogantrabbit.com (da9:mpr0:0:25:0): Retrying command, 3 more tries remain
/var/log/messages.0.bz2-Jul 10 23:39:54 truenas.arrogantrabbit.com (da9:mpr0:0:25:0): WRITE(10). CDB: 2a 00 02 00 02 a0 00 00 e0 00
/var/log/messages.0.bz2:Jul 10 23:39:54 truenas.arrogantrabbit.com (da0:mpr0:0:16:0): CAM status: CCB request completed with an error
/var/log/messages.0.bz2-Jul 10 23:39:54 truenas.arrogantrabbit.com (da0:mpr0:0:16:0): Retrying command, 3 more tries remain
/var/log/messages.0.bz2-Jul 10 23:39:54 truenas.arrogantrabbit.com (da0:mpr0:0:16:0): WRITE(10). CDB: 2a 00 02 00 02 a0 00 00 e0 00
/var/log/messages.0.bz2:Jul 10 23:39:54 truenas.arrogantrabbit.com (da0:mpr0:0:16:0): CAM status: CCB request completed with an error
/var/log/messages.0.bz2-Jul 10 23:39:54 truenas.arrogantrabbit.com (da0:mpr0:0:16:0): Retrying command, 3 more tries remain
/var/log/messages.0.bz2-Jul 10 23:39:54 truenas.arrogantrabbit.com (da8:mpr0:0:24:0): WRITE(16). CDB: 8a 00 00 00 00 08 2f 7f fc a0 00 00 00 e0 00 00
/var/log/messages.0.bz2:Jul 10 23:39:54 truenas.arrogantrabbit.com (da8:mpr0:0:24:0): CAM status: CCB request completed with an error
/var/log/messages.0.bz2:Jul 10 23:39:54 truenas.arrogantrabbit.com (da9:mpr0:0:25:0): CAM status: CCB request completed with an error
/var/log/messages.0.bz2-Jul 10 23:39:54 truenas.arrogantrabbit.com (da9:mpr0:0:25:0): Retrying command, 3 more tries remain
/var/log/messages.0.bz2-Jul 10 23:39:54 truenas.arrogantrabbit.com (da5:mpr0:0:21:0): WRITE(16). CDB: 8a 00 00 00 00 08 2f 7f fc a0 00 00 00 e0 00 00
/var/log/messages.0.bz2-Jul 10 23:39:54 truenas.arrogantrabbit.com (da8:mpr0:0:24:0): Retrying command, 3 more tries remain
/var/log/messages.0.bz2-Jul 10 23:39:54 truenas.arrogantrabbit.com (da8:mpr0:0:24:0): WRITE(10). CDB: 2a 00 02 00 02 a0 00 00 e0 00
/var/log/messages.0.bz2:Jul 10 23:39:54 truenas.arrogantrabbit.com (da5:mpr0:0:21:0): CAM status: CCB request completed with an error
/var/log/messages.0.bz2-Jul 10 23:39:54 truenas.arrogantrabbit.com (da5:mpr0:0:21:0): Retrying command, 3 more tries remain
/var/log/messages.0.bz2:Jul 10 23:39:54 truenas.arrogantrabbit.com (da8:mpr0:0:24:0): CAM status: CCB request completed with an error
/var/log/messages.0.bz2-Jul 10 23:39:54 truenas.arrogantrabbit.com (da8:mpr0:0:24:0): Retrying command, 3 more tries remain
/var/log/messages.0.bz2-Jul 10 23:39:54 truenas.arrogantrabbit.com (da2:mpr0:0:18:0): WRITE(16). CDB: 8a 00 00 00 00 08 2f 7f fc a0 00 00 00 e0 00 00
/var/log/messages.0.bz2-Jul 10 23:39:54 truenas.arrogantrabbit.com (da5:mpr0:0:21:0): WRITE(10). CDB: 2a 00 02 00 02 a0 00 00 e0 00
/var/log/messages.0.bz2:Jul 10 23:39:54 truenas.arrogantrabbit.com (da5:mpr0:0:21:0): CAM status: CCB request completed with an error
/var/log/messages.0.bz2-Jul 10 23:39:54 truenas.arrogantrabbit.com (da5:mpr0:0:21:0): Retrying command, 3 more tries remain
/var/log/messages.0.bz2-Jul 10 23:39:54 truenas.arrogantrabbit.com (da11:mpr0:0:27:0): WRITE(16). CDB: 8a 00 00 00 00 08 2f 7f fc a0 00 00 00 e0 00 00
/var/log/messages.0.bz2:Jul 10 23:39:54 truenas.arrogantrabbit.com (da11:mpr0:0:27:0): CAM status: CCB request completed with an error
/var/log/messages.0.bz2:Jul 10 23:39:54 truenas.arrogantrabbit.com (da2:mpr0:0:18:0): CAM status: CCB request completed with an error
/var/log/messages.0.bz2-Jul 10 23:39:54 truenas.arrogantrabbit.com (da2:mpr0:0:18:0): Retrying command, 3 more tries remain
/var/log/messages.0.bz2-Jul 10 23:39:54 truenas.arrogantrabbit.com (da2:mpr0:0:18:0): WRITE(10). CDB: 2a 00 02 00 02 a0 00 00 e0 00
/var/log/messages.0.bz2:Jul 10 23:39:54 truenas.arrogantrabbit.com (da2:mpr0:0:18:0): CAM status: CCB request completed with an error
/var/log/messages.0.bz2-Jul 10 23:39:54 truenas.arrogantrabbit.com (da2:mpr0:0:18:0): Retrying command, 3 more tries remain
/var/log/messages.0.bz2-Jul 10 23:39:54 truenas.arrogantrabbit.com (da0:mpr0:0:16:0): WRITE(10). CDB: 2a 00 02 00 02 a0 00 00 e0 00
/var/log/messages.0.bz2-Jul 10 23:39:54 truenas.arrogantrabbit.com (da11:mpr0:0:27:0): Retrying command, 3 more tries remain
/var/log/messages.0.bz2-Jul 10 23:39:54 truenas.arrogantrabbit.com (da11:mpr0:0:27:0): WRITE(10). CDB: 2a 00 02 00 02 a0 00 00 e0 00
/var/log/messages.0.bz2:Jul 10 23:39:54 truenas.arrogantrabbit.com (da0:mpr0:0:16:0): CAM status: SCSI Status Error
/var/log/messages.0.bz2-Jul 10 23:39:54 truenas.arrogantrabbit.com (da0:mpr0:0:16:0): SCSI status: Check Condition
/var/log/messages.0.bz2-Jul 10 23:39:54 truenas.arrogantrabbit.com (da0:mpr0:0:16:0): SCSI sense: UNIT ATTENTION asc:29,0 (Power on, reset, or bus device reset occurred)
/var/log/messages.0.bz2:Jul 10 23:39:54 truenas.arrogantrabbit.com (da11:mpr0:0:27:0): CAM status: CCB request completed with an error
/var/log/messages.0.bz2-Jul 10 23:39:54 truenas.arrogantrabbit.com (da11:mpr0:0:27:0): Retrying command, 3 more tries remain
/var/log/messages.0.bz2-Jul 10 23:39:54 truenas.arrogantrabbit.com (da3:mpr0:0:19:0): WRITE(10). CDB: 2a 00 02 00 02 a0 00 00 e0 00
/var/log/messages.0.bz2-Jul 10 23:39:54 truenas.arrogantrabbit.com (da0:mpr0:0:16:0): Retrying command (per sense data)
/var/log/messages.0.bz2:Jul 10 23:39:54 truenas.arrogantrabbit.com (da3:mpr0:0:19:0): CAM status: SCSI Status Error
/var/log/messages.0.bz2-Jul 10 23:39:54 truenas.arrogantrabbit.com (da3:mpr0:0:19:0): SCSI status: Check Condition
/var/log/messages.0.bz2-Jul 10 23:39:54 truenas.arrogantrabbit.com (da3:mpr0:0:19:0): SCSI sense: UNIT ATTENTION asc:29,0 (Power on, reset, or bus device reset occurred)
/var/log/messages.0.bz2-Jul 10 23:39:54 truenas.arrogantrabbit.com (da3:mpr0:0:19:0): Retrying command (per sense data)
/var/log/messages.0.bz2-Jul 10 23:39:54 truenas.arrogantrabbit.com (da6:mpr0:0:22:0): WRITE(10). CDB: 2a 00 02 00 02 a0 00 00 e0 00
/var/log/messages.0.bz2:Jul 10 23:39:54 truenas.arrogantrabbit.com (da6:mpr0:0:22:0): CAM status: SCSI Status Error
/var/log/messages.0.bz2-Jul 10 23:39:54 truenas.arrogantrabbit.com (da6:mpr0:0:22:0): SCSI status: Check Condition
/var/log/messages.0.bz2-Jul 10 23:39:54 truenas.arrogantrabbit.com (da6:mpr0:0:22:0): SCSI sense: UNIT ATTENTION asc:29,0 (Power on, reset, or bus device reset occurred)
/var/log/messages.0.bz2-Jul 10 23:39:54 truenas.arrogantrabbit.com (da5:mpr0:0:21:0): WRITE(10). CDB: 2a 00 02 00 02 a0 00 00 e0 00
/var/log/messages.0.bz2-Jul 10 23:39:54 truenas.arrogantrabbit.com (da6:mpr0:0:22:0): Retrying command (per sense data)
/var/log/messages.0.bz2:Jul 10 23:39:54 truenas.arrogantrabbit.com (da5:mpr0:0:21:0): CAM status: SCSI Status Error
/var/log/messages.0.bz2-Jul 10 23:39:54 truenas.arrogantrabbit.com (da5:mpr0:0:21:0): SCSI status: Check Condition
/var/log/messages.0.bz2-Jul 10 23:39:54 truenas.arrogantrabbit.com (da5:mpr0:0:21:0): SCSI sense: UNIT ATTENTION asc:29,0 (Power on, reset, or bus device reset occurred)
/var/log/messages.0.bz2-Jul 10 23:39:54 truenas.arrogantrabbit.com (da5:mpr0:0:21:0): Retrying command (per sense data)
/var/log/messages.0.bz2-Jul 10 23:39:54 truenas.arrogantrabbit.com (da8:mpr0:0:24:0): WRITE(10). CDB: 2a 00 02 00 02 a0 00 00 e0 00
/var/log/messages.0.bz2:Jul 10 23:39:54 truenas.arrogantrabbit.com (da8:mpr0:0:24:0): CAM status: SCSI Status Error
/var/log/messages.0.bz2-Jul 10 23:39:54 truenas.arrogantrabbit.com (da8:mpr0:0:24:0): SCSI status: Check Condition
/var/log/messages.0.bz2-Jul 10 23:39:54 truenas.arrogantrabbit.com (da8:mpr0:0:24:0): SCSI sense: UNIT ATTENTION asc:29,0 (Power on, reset, or bus device reset occurred)
So either the cable toast, or the HBA toast.
I’ve ordered another cable (will arrive today) and updated firmware on the HBA yesterday, and will try again tonight. Hope it’s a cable, those look like signal integrity issues. But who knows.
It was the cable - replacement cable works fine. I’ve opened the ebay return for the defective one, ebay issued me a label, and hillarity ensued. The length the dude with questionable ethics went to get me to cancel the return request is mindboggling. (I shoudl have known somethign was fishy – the dude’s profile is in china, USA on the listing, and Canadian flag on the avatar… )
Message log for some entertainment value
Me:
Hi Noah,
The cable turns out to be defective. See below what led me to the conclusion. I’m going to open a return, please destroy it once you get it, don’t re-sell it, it’s defective.I am replacing my 3008-series LSI HBA with 3808 series one. Because they have different connectors on the board I needed a new cable to connect to backplane. The backplane does not change, it’s a supermicro singel expander backplane.
As soon as I connected both HBA and the cable you sent me – I immediately started seeing data integrity retries like these:
<snippet of the log>
Usually these are symptoms of a bad cable, but because HBA is also new it could have been bad too. So I ordered another cable on Amazon (hence the delay). Using that cable, and the new HBA I dont’ see these errors.
This confirms the cable is bad.
Them:
Hi there, thank you for your message. Although no other customers have reported issues with this product since it was launched, based on your description, we apologize if our product may have caused you an unpleasant shopping experience.
Friend, please don’t return the item just yet. I will check with our product department to confirm if the problem is with the cable. I’ll get back to you as soon as we have a confirmation. Thank you.
Regards,
Noah
Me:
Swapping the cable with another removes the issue. That’s the evidence that’s the culprit was the cable.
Not your fault, you could not have known, you can’t possibly test every single cable, and some small percentage of any mass produced product can be faulty. Again, not your fault, no need to apologize.
Them:
Hi friend,
Our product team has confirmed that your card supports NVMe, but our product is only compatible with SAS and does not support NVMe.
Here’s the solution we’d like to propose: you can keep the item and handle it as you wish. If you could cancel the return request, we’ll process a full refund for you right away.
Would that work for you? Thank you.
Me:
Hi Noah. This explanation does not make any sense. I have no NVME devices in the system. Backplane communicates with the HBA via SAS. The fact that HBA can support NVME is pretty irrelevant.
The “UNIT ATTENTION” and “SCSI Status Error” are classic signs of a physical layer problem, bad connection, or a fauly cable. This also confirmed by me using a similar cable from amazon that works just fine.
I’m convinced the cable is defectiive.Yes, I can cancel the return request and destroy the cable myself.
To ensure the process goes smoothly, please issue the full refund first. As soon as I receive confirmation that the refund has been processed and is in my account, I will immediately go in and cancel the return request.
Them:
Hi friend,
Thank you again for your understanding and the information.
This should be the root of the problem. Could you please cancel the return request? We’ll process a refund for you right away.
Would that work for you?
Me:
Noah,
Once I close return request the case is closed and I will no loinger be able to open another for the same item.
Since you are not willing to send refund first, let’s follow proper ebay procedures and I’m going to mail the cable back tomorrow.
Since you say no other customers report these issue a single defect don’t affect your seller account.
Them:
Hi friend, It won’t affect the account, but it will affect the product listing. The return reason you provided might lead to this listing being taken down. To keep it available for sale, we hope you could cancel the return request first, and we’ll arrange the refund for you right away.
We’ve been in business for 8 years, so please trust that we’ll process the refund promptly once you cancel the request. Thank you.
Me:
No, I don’t trust you. I don’t know you.
Ebay won’t take down the product due to one defective return. You shoudl know that since you’ve been in business for 8 years.
They might take down the listing if it receives a lot of returns. And in this case, it indeed deserves to be taken down.
I suggested you send the refund first twice already, but you are not willing to do so. The only reason I see that you are not going to. Once I close the return request I have no leverage.
So my offer is off. I don’t like that you are tryign to circumvent ebay polciies and protectiosn. I don’t like that you tried to shift blame to “nvme”.
I’m not going to be cancelling a return. Lets play by the rules of the platform we are both using, for our both protection.
Them:
Alright then, friend. My original intention was to resolve the issue. I’ll process the refund for you, and we’ll also take down the listing. Once you receive the refund, could you please cancel the return request? Thank you.
My last message:
Why would you take down the listing of a perfectly sound product?
I’m not cancelling anything. I’m sending back the cable. I’m folloiwnig ebay policies to the T.
Their last rambling response:
Hi friend.
If this product keeps having issues, we will still continue to sell it. Please believe that my original intention is to solve the problem. If we can keep the listing active, why not communicate with you about canceling the return request first? As I mentioned earlier, the reason for the return request will be regarded as a problematic transaction. What if it’s an incompatibility issue rather than a problem with the product itself?
I’m really sorry for the unpleasant shopping experience. If possible, we still hope you won’t return the item. We won’t resell the returned product, and we’ll also bear the return shipping cost. I’ll process the refund for you now, okay? Sincerely many apologies again.
Regards Noah
I’m selling on ebay for 20+ years, and can’t stand this sort of disgusting behaviour that ruins ebay for everyone.
More importantly, power indeed dropped another 9 or so watts (likely more, right now time machine is doing some scheniningans, I’ll retest later, but it’s quite promising as it is)
That is baffling shenanigans. I’ve had vendors of cheap stuff simply refund a purchase price instead of doing a return (especially shipping to china)… And I’ve had vendors go through bribes to get their review scores improved
… but trying to bypass a return process seems silly, I don’t even see why it would be a ding against them, as long as they actually handle the return or refund.
The return reason (item defective) falls under umbrella of SNAD (significantly not as described), and it’s a defect on their account. High rate of defects totally affect their account — from higher fees to bans (despite of what the dude said). Amount of manipulative bullshit to get me to close the returns strongly hints he does get high return rate. And to the lesser degree — return shipping likely (definitely) exceeds the value of the cable.
Either way, they could have refunded without me having to close the return first — because if they are going to refund anyway that makes no difference. Strong resistance there means they not only wanted to return to go away but also scam me and never refund at all — after all, I voluntarily closed the return, I’m a happy customer.
I reported his account and the message as soon as he busted out the “cancel the return” spiel. I strongly recommend reporting these sellers if you encounter one. eBay will investigate and take action, more likely of there is a lot of reports. This is the only way to combat this behavior.
1 Like
Out of curiosity, couldn’t you just request a chargeback?
eBay provides their own buyer protection, that is often better than that of the banks, and definitely faster. Furthermore, technically, I’m paying eBay, and eBay (so far) did nothing wrong, so chargeback would not be appropriate.
I did not need to engage with the dude at all, I just gave him heads up out of sheer compassion. Ebay lets me print the return label (at the dude’s expense, regardless of what they think about it), and if they don’t refund within 3 days of receipt – eBay will refund, charge the seller, and ding them even more. Nobody wants that, so this almost never gets to that point. The dude knows that perfectly well, he just wanted to improve his stats – talk is cheap, and some buyers may cave.
It’s my favorite platform – it’s safe for buyers and fair for sellers. I’m both, and I appreciate both sides. So far – top notch experience for 20 years and counting. (Amazon on the other hand… I despise deeply… and yet I’m paying for prime)
This interaction was more entertaining than concerning – it’s nice to be in the position where you know you are right and can see though their weak attempts at scams.
1 Like