C'DaleRider
06-15-2007, 05:53 AM
Kyle at {H} just did another review of an OCZ GameXstream power supply....this their newest 1010W unit. Not surprisingly, it too exhibited the significant and out of spec ripple problems that plague the Epsilon line of power supplies......not to mention the 2 units Kyle had refused to put out more than 850W no matter how the demand was spread across any and all rails.
The +3.3V and +5V rails were over spec on ripple at just 50% of rated output. The+12V also went way out of ATX spec on ripple, too. Of course, OCZ fudged the ATX spec by listing the ripple on the +12V rail as being "No Greater Than 140mV"......which is unfortunately well outside ATX spec of max ripple allowed (120mV).
So, when will OCZ finally dump FSP Epsilons as their OEM source for the GameXstream power supplies? Every legitimate test of any GameXstream....{H}, JonnyGuru among others, all note out-of-spec ripple on all the rails. And if you like your components, ripple can shorten their lives significantly. Ture, not in a few months, but a year or so down the line and the out-of-spec ripple can end the life of components prematurely.
"Overall the load testing portion of the GameXStream 1010w was disappointing. It seems that the GameXStream 1010w is pushing the DC output capacity of this unit beyond its limits when tested at our more relevant operating temperature of 45c. The units fan is certainly capable of cooling the unit as the highest exhaust temperature recorded was below 50c with a 45c input. This coupled with the units good efficiency make it unlikely that the unit was overheating and was rather simply pushed too far in labeling the DC output.
"...the GameXStream 1010w was like the other members of the Epsilon line by starting off active and being almost out of specifications at only 50% load. At 75% load the ripple/noise of the unit was out of specification on the 3.3v/5v rails. This does not bode well for a system as system memory is powered off of the 3.3v. It has additionally been reported by other reviewers that at high loads the 12v rails were out of specification as well and indeed the unit even advertises the fact that it does not meet the ATX12v allowance of 120mV instead it is specified as being 140mV. Since our unit was unable to complete Test #4 we can not completely verify the 12v figures, however it should be noted that the 12v1 was around 100mV at both Test #3 and the Torture Test. Typically noise/ripple increase with load and I would expect this trend to continue with the OCZ GameXStream 1010w."
"The OCZ GameXStream 1010w entered testing with decent build quality from a well known OEM, but from a product line that has had a reputation for not being their best to date. However, our previous experience with the lower wattage Epsilon based FSP BlueStorm II was very good so we were hopeful that the GameXStream would follow in its foot steps. Unfortunately, that was not to be as the unit had a number of failings. The first and most apparent was apparent right out of the box. The unit is labeled with 2 different 12v amperage ratings which were quickly followed by the unit being specified for more ripple/noise than what is allowed by the ATX12v specification. Yes, the unit does not comply with ATX12v spec by OCZ’s own admission. We recently saw this issue in our 450w-500w Battle Royal from a couple of less-than-reputable manufacturers, so finding this issue carrying over to a much better OEM and brand is extremely disappointing. This worry was then borne out during testing when the OCZ GameXStream exhibited more ripple/noise on the 3.3v/5v rails than is allowed by the ATX12v specification. Previously other reviewers have reported out of specification ripple/noise on the 12v rails with other OCZ/FSP Epsilon based units at high loads, however we were unable to verify these ourselves because neither unit we had in for testing would complete a 100% load test. As it stands today the OCZ GameXStream 1010w is not all it should or could be. If OCZ can get together with FSP and iron out the 12v amperage and ripple/noise ratings issue, the noise/ripple performance issue, and full power loading at relevant temperatures (not 25c)"
The +3.3V and +5V rails were over spec on ripple at just 50% of rated output. The+12V also went way out of ATX spec on ripple, too. Of course, OCZ fudged the ATX spec by listing the ripple on the +12V rail as being "No Greater Than 140mV"......which is unfortunately well outside ATX spec of max ripple allowed (120mV).
So, when will OCZ finally dump FSP Epsilons as their OEM source for the GameXstream power supplies? Every legitimate test of any GameXstream....{H}, JonnyGuru among others, all note out-of-spec ripple on all the rails. And if you like your components, ripple can shorten their lives significantly. Ture, not in a few months, but a year or so down the line and the out-of-spec ripple can end the life of components prematurely.
"Overall the load testing portion of the GameXStream 1010w was disappointing. It seems that the GameXStream 1010w is pushing the DC output capacity of this unit beyond its limits when tested at our more relevant operating temperature of 45c. The units fan is certainly capable of cooling the unit as the highest exhaust temperature recorded was below 50c with a 45c input. This coupled with the units good efficiency make it unlikely that the unit was overheating and was rather simply pushed too far in labeling the DC output.
"...the GameXStream 1010w was like the other members of the Epsilon line by starting off active and being almost out of specifications at only 50% load. At 75% load the ripple/noise of the unit was out of specification on the 3.3v/5v rails. This does not bode well for a system as system memory is powered off of the 3.3v. It has additionally been reported by other reviewers that at high loads the 12v rails were out of specification as well and indeed the unit even advertises the fact that it does not meet the ATX12v allowance of 120mV instead it is specified as being 140mV. Since our unit was unable to complete Test #4 we can not completely verify the 12v figures, however it should be noted that the 12v1 was around 100mV at both Test #3 and the Torture Test. Typically noise/ripple increase with load and I would expect this trend to continue with the OCZ GameXStream 1010w."
"The OCZ GameXStream 1010w entered testing with decent build quality from a well known OEM, but from a product line that has had a reputation for not being their best to date. However, our previous experience with the lower wattage Epsilon based FSP BlueStorm II was very good so we were hopeful that the GameXStream would follow in its foot steps. Unfortunately, that was not to be as the unit had a number of failings. The first and most apparent was apparent right out of the box. The unit is labeled with 2 different 12v amperage ratings which were quickly followed by the unit being specified for more ripple/noise than what is allowed by the ATX12v specification. Yes, the unit does not comply with ATX12v spec by OCZ’s own admission. We recently saw this issue in our 450w-500w Battle Royal from a couple of less-than-reputable manufacturers, so finding this issue carrying over to a much better OEM and brand is extremely disappointing. This worry was then borne out during testing when the OCZ GameXStream exhibited more ripple/noise on the 3.3v/5v rails than is allowed by the ATX12v specification. Previously other reviewers have reported out of specification ripple/noise on the 12v rails with other OCZ/FSP Epsilon based units at high loads, however we were unable to verify these ourselves because neither unit we had in for testing would complete a 100% load test. As it stands today the OCZ GameXStream 1010w is not all it should or could be. If OCZ can get together with FSP and iron out the 12v amperage and ripple/noise ratings issue, the noise/ripple performance issue, and full power loading at relevant temperatures (not 25c)"